--- title: 'DESI BAO: Precision Cosmology Results' url: https://www.emergentmind.com/topics/desi-bao-results type: topic --- # DESI BAO: Precision Cosmology Results The Dark Energy Spectroscopic Instrument (DESI) Baryon Acoustic Oscillation (BAO) results represent a major advance in low-redshift cosmological calibration and precision measurement of the cosmic expansion history. By leveraging a multi-tracer, wide-redshift survey of galaxies and quasars, DESI has provided sub-percent accuracy BAO distances and enabled stringent tests of the standard $\Lambda$CDM model, dark energy equation-of-state dynamics, and the Hubble constant ($H_0$) tension. The following sections detail the scope, methodology, main empirical findings, and theoretical impact of the DESI BAO results as established in recent literature. ## 1. Tracer Samples and Measurement Strategy DESI's first-year BAO dataset comprises seven principal measurements, spanning redshifts $z\simeq0.3$ to $z\simeq2.3$. The sampled tracers include the Bright Galaxy Sample (BGS, $z\approx0.3$), Luminous Red Galaxies (LRG1/2, $z\approx0.51,\,0.71$), Emission-Line Galaxies (ELG, $z\approx1.3$), Quasars (QSO, $z\approx1.49$), and the Lyman-$\alpha$ forest ($z\approx2.33$) [2404.03000, 2411.14189]. Distances are reported as ratios to the comoving sound horizon at the drag epoch, $r_d$, for each of three standard measures: - $D_M(z)$: Transverse comoving (angular diameter) distance. - $D_H(z)$: Hubble (radial) distance, $c/H(z)$. - $D_V(z)$: Spherically averaged BAO distance, $[z\,D_M^2(z)\,D_H(z)]^{1/3}$. Each measurement includes statistical and systematic uncertainties (typical fractional errors: $\sim$1--3%) and is encoded in a block-diagonal covariance matrix, allowing for inter-dependence between $D_M$ and $D_H$ at each $z$ [2411.14189]. ## 2. BAO Likelihood Construction and Cosmological Model Dependencies DESI BAO measurements are statistically incorporated via a joint $\chi^2$ likelihood: $$ \chi^2_{\mathrm{BAO}} = \sum_{i,j} [O_i^{\mathrm{obs}} - O_i^{\mathrm{th}}]\, (C^{-1}_{\mathrm{BAO}})_{ij}\, [O_j^{\mathrm{obs}} - O_j^{\mathrm{th}}], $$ where $O_i \in \{ D_V/r_d,\, D_M/r_d,\, D_H/r_d \}$ at each redshift, and $C_{\mathrm{BAO}}$ is the published covariance [2411.14189, 2404.03000]. Cosmological predictions employ background expansion in flat $\Lambda$CDM: $$ H(z) = H_0\, \sqrt{ \Omega_m (1+z)^3 + (1-\Omega_m) }, $$ with observables built from theoretical integrals out to each effective $z$ and using $r_d$ calculated from pre-recombination physics. ## 3. Impact on $H_0$ and Hubble Tension One of DESI's crucial achievements is in improving constraints on $H_0$ in joint fits with CMB data. When combined with non-Planck CMB datasets (WMAP, ACT, SPT), DESI BAO results produce: - WMAP + DESI BAO: $H_0 = 68.86 \pm 0.68\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ - WMAP + ACT + DESI BAO: $H_0 = 68.72 \pm 0.51\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ - WMAP + SPT + DESI BAO: $H_0 = 68.62 \pm 0.52\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ These are in $3.4\sigma$--$3.8\sigma$ tension with the SH0ES local ladder $H_0=73.04\pm1.04\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$, about $1\sigma$ lower in significance than Planck+DESI BAO combinations [2411.14189]. Inclusion of DESI BAO shifts $H_0$ upward by $\sim$1 km/s/Mpc and reduces random error by $\sim$40% relative to pre-DESI BAO catalogues (e.g., SDSS DR7/DR16, 6dFGS), thereby reducing the tension. A purely data-driven analysis, circumventing $r_d$-external calibrations, yields $H_0=68.4^{+1.0}_{-0.8}\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ (1.3% precision), fully consistent with Planck and TRGB determinations but $4.3\sigma$ lower than SH0ES [2412.13045]. ## 4. BAO-driven Constraints on Dark Energy Dynamics Several analyses employing the DESI BAO dataset alone or in combination with CMB/SN Ia data have explored the parameter space of dynamical dark energy: - In $w_0w_a$CDM (Chevallier–Polarski–Linder parametrization), DESI BAO alone yields $w_0 = -0.54^{+0.38}_{-0.21}$, $w_a = -1.66^{+0.43}_{-1.30}$, and $\Omega_m = 0.345^{+0.044}_{-0.025}$. - Adding CMB and SN samples provides stronger evidence for $w_0 > -1$ and $w_a < 0$, with joint analyses producing best fits as far as $w_0 = -0.73\pm0.07$, $w_a = -1.01\pm0.30$ [2507.22575, 2412.04830]. Analysis of BAO information criteria (AIC/BIC) shows modest statistical support for dynamically evolving dark energy, particularly in two-parameter forms like Barboza–Alcaniz (BA) and FSLL, though $\Lambda$CDM remains competitive when LRG1/LRG2 measurements are excluded [2412.04830]. The monopole components (angle-averaged distances $D_V$) of LRG1/2 at $z=0.51,\,0.71$ are disproportionately responsible for pushing best-fit $(w_0, w_a)$ away from $(-1, 0)$; exclusion of LRG2 monopole largely restores $\Lambda$CDM consistency [2405.02168]. ## 5. Model-independent Expansion History and Consistency Tests Non-parametric Gaussian-process and crossing-statistics reconstructions using DESI BAO: - Indicate mild evidence ($\sim2$–$3\sigma$) for evolving $H(z)$, weaker present-day acceleration (deceleration parameter $q(0)\approx -0.3$), and time-varying $\mathcal{O}_\mathrm{m}(z)$ inconsistent with constant $\Omega_m$ [2408.04432, 2405.19178, 2405.04216]. - However, results depend significantly on SN Ia sample selection (e.g., DES-5YR vs. PantheonPlus or Union3). PantheonPlus and Union3 fits are fully consistent with $\Lambda$CDM at $1$–$2\sigma$, while DES-5YR pushes reconstructed quantities outside $3\sigma$ bands, requiring further investigation of SN systematics [2405.19178]. - Joint DESI+SDSS BAO datasets restore expansion histories to full consistency with the Planck benchmark; the two surveys individual reconstructions disagree at low-$z$, highlighting the need for cross-survey systematics control [2408.04432]. ## 6. Statistical Robustness, Systematics, and Multi-Tracer Optimization DESI BAO analyses employ sophisticated blinding, reconstruction (e.g., "RecSym"), and template marginalization protocols [2404.03000, 2402.14070]. Systematic errors from nonlinear clustering, RSD, template fitting, and sample-variance suppression have been quantified and found to produce biases well below statistical precision (typically $<$0.1% for isotropic, $<$0.2% for anisotropic measurements). Multi-tracer approaches, particularly for overlapping LRG+ELG galaxies at $0.8