Reanalyzing DESI DR1: 5. Cosmological Constraints with Simulation-Based Priors

Abstract

We analyze the public DESI full-shape clustering data using simulation-based priors (SBPs). Our priors are obtained by fitting normalizing flows to the distribution of EFT parameters measured from field-level simulations, themselves generated using tailored halo occupation distribution (HOD) models for each tracer. Incorporating SBPs in a power spectrum analysis significantly enhances $Λ$CDM cosmological parameter constraints; in combination with BAO information from DESI DR2 and a BBN prior on the baryon density, we find the matter density parameter $Ω_m=0.2987\pm 0.0066$, the Hubble constant $H_0=68.80\pm 0.35\,\mathrm{km}\,\mathrm{s}^{-1}\mathrm{Mpc}^{-1}$, and the mass fluctuation amplitude $σ_8 = 0.766\pm 0.015$ (or the lensing parameter $S_8=0.764\pm 0.018$), which are $1\%$, $40\%$ and $50\%$ stronger than the baseline results, though with a notable downwards shift in $σ_8$. The SBPs also have a significant impact in extended models, with the dark energy figure-of-merit improving by $70\%$ ($20\%$) in a $w_0w_a$CDM analysis when combining with the CMB (and supernovae). In the SBP analysis, we do not find statistically significant evidence for dynamical dark energy: the equation of state parameters are consistent with a cosmological constant within $2.2σ$ ($1.4σ$) in analyses without (with) supernovae. The neutrino mass constraints are also enhanced, with the $95\%$ limits $M_ν<0.073\,\mathrm{eV}$ and $M_ν<0.090\,\mathrm{eV}$ in $Λ$CDM and $w_0w_a$CDM respectively. The latter is the strongest constraint obtained to date and reinforces the preference for the normal neutrino mass hierarchy, regardless of the background dynamics. While our results are sensitive to HOD modeling assumptions, they clearly demonstrate that the inclusion of small-scale information can significantly sharpen cosmological parameter constraints.

Figures (8)

• Figure 1: $\Lambda$CDM constraints: Posterior distributions for the $\Lambda{\rm CDM}$ parameters $\Omega$, $H_0$, $\sigma_8$ and $S_8$ obtained from different combinations of datasets: the DESI DR1 redshift-space power spectrum and bispectrum monopole ($P_\ell$ + $B_0$), expansion history measurements from DESI DR2 (BAO) and the Planck temperature and polarization anisotropy data, including lensing reconstruction from both the Plancknpipe PR4 maps and ACT DR6 (CMB). Results are shown for conservative Bayesian priors (CBP, dashed) and simulation-based priors (SBP, solid). Since the addition of higher-order statistics does not alter the cosmological and EFT parameter constraints, we do not include the DESI DR1 galaxy bispectrum when including simulation-based priors; this is validated in Appendix \ref{['app:bisp']}. The posteriors shrink significantly when SBPs are implemented, with a notable shift to lower values of $\sigma_8$ and $S_8$, which may indicate systematic effects or overly confident priors. The main parameter constraints derived from the CMB and DESI full-shape-plus-BAO datasets (using SBPs for the latter) are broadly consistent at the $2.8\sigma$ level, whilst mean values of $S_8$ differ between the two by $3.2\sigma$.
• Figure 2: Bias parameters: Posteriors on linear and quadratic bias parameters from $P_\ell$ and $P_\ell+B_0$ analyses using conservative Bayesian priors (CBP), and from $P_\ell$ analyses using simulation-based priors (SBP). The lower triangles show the bias parameters for the LRG1, LRG2, LRG3 galaxy samples, while the upper triangle presents those for BGS, ELG and QSO samples; these are represented by solid and dashed lines in the diagonal plots respectively. Superscripts (1)-(6) correspond to BGS, LRG1, LRG2, LRG3, ELG2, and QSO samples, respectively. Incorporating SBPs dramatically shrinks the posterior volume and pulls the bias parameters into closer agreement with the results of the standard $P_\ell+B_\ell$ analysis adopting CBPs, though there are some slight anomalies such as for $b_1^{(2)}$. We note that the SBPs of this work apply also to the counterterm and stochasticity parameters, which are not shown above.
• Figure 3: Dynamical dark energy: Two-dimensional posterior distributions on the dynamical dark energy parameters for various combinations of datasets, as indicated by the captions. We show results with both conservative and simulation-based priors (CBPs & SBPs). Left panel: The use of SBPs reduces the preference for dynamical dark energy from $2.8\sigma$ to $2.2\sigma$ without relying on supernova distance information (see Tab. \ref{['tab:chi2']} for frequentist constraints) Right panel: When supernova data are included, adopting SBPs alleviates the preference for a time-evolving dark energy equation of state in the conventional EFT-based analysis from $2.9\sigma$ to $1.4\sigma$.
• Figure 4: $\Omega_m$ constraints: Marginalized one-dimensional posterior on $\Omega_m$, when fixing the background model to $w_0w_a{\rm CDM}$. We also show the $\Omega_m$ posterior from the official DESI analysis of the BAO DR2 data in the $\Lambda{\rm CDM}$ model (black dot-dashed line). The addition of the DESI DR1 FS information to the CMB and BAO dataset tightens the $\Omega_m$ posterior by $11\%$ when using conservative Bayesian priors (CBP) and by $31\%$ when adopting simulation-based priors (SBP) and shifts the mean towards lower values in a closer agreement with the DESI-only $\Lambda{\rm CDM}$ result, without relying on supernova information.
• Figure 5: Massive neutrinos: One-dimensional marginalized posterior distributions for the sum of neutrino masses when fixing the background model to $\Lambda{\rm CDM}$ ( left panel) and $w_0w_a{\rm CDM}$ ( right panel). Left panel:$M_\nu+\Lambda{\rm CDM}$ analyses using the DESI DR2 BAO measurements, the DESI DR1 power spectrum $P_\ell$ and bispectrum $B_0$, and Planck CMB data plus Planck and ACT lensing (red). Right panel: Combined constraints on neutrino masses in the $M_\nu+w_0w_a{\rm CDM}$ analyses using the conservative EFT priors (dashed) and SBPs (solid).