Baryon Acoustic Oscillations from the C IV Forest with DESI DR2

TL;DR

This work demonstrates BAO measurements using the CIV forest cross-correlated with DESI DR2 tracers, extending BAO analyses to higher redshift regimes by leveraging CIV absorption alongside QSOs and ELGs. Using two CIV side-bands and continuum fitting with picca, the authors measure isotropic BAO via $\alpha_{\rm iso}$ and convert it to $D_{ m V}/r_{ m d}$ at effective redshifts $z_{\rm eff}\approx 1.92$ and $1.47$, achieving a $4.2\sigma$ detection for CIV$\times$QSO and a $2.5\sigma$ detection for CIV$\times$ELG. They model metal contamination from SiII(1527) and FeII(1608), validate results with multiple analysis variations and a data-split test, and find results consistent with the standard $\Lambda$CDM cosmology, highlighting CIV as a viable high-redshift BAO tracer. The study also outlines paths for improvements (e.g., including BAL quasars, more repeated quasar observations, and alternative continuum estimators) to tighten future DESI BAO constraints. Overall, this constitutes the first DESI BAO measurement using the CIV forest cross-correlated with QSOs and a proof-of-concept with ELGs, expanding the toolkit for precision cosmology at $z\sim 1.5$--$2$.

Abstract

We present a measurement of Baryon Acoustic Oscillations (BAO) in the cross-correlation of triply ionized carbon C IV absorption with the positions of quasars (QSO) and Emission Line Galaxies (ELG). We use quasars and ELGs from the second data release (DR2) of the Dark Energy Spectroscopic Instrument (DESI) survey. Our data sample consists of 2.5 million quasars, 3.1 million ELGs, and the C IV absorption is measured along the line of sight of 1.5 million high redshift quasars with $z > 1.3$. We measure the isotropic BAO signal at 4.2$σ$ for the CIV$\times$QSO cross-correlation. This translates into a 3.0% precision measurement of the ratio of the isotropic distance scale, $D_{\rm V}$, and the sound horizon at the drag epoch, $r_{\rm d}$, with $D_{\rm V}/r_{\rm d}(z_{\rm eff} = 1.92) = 30.3 \pm 0.9$. We make the first detection of the BAO feature in the CIV$\times$ELG cross-correlation at a significance of 2.5$σ$ and find $D_{\rm V}/r_{\rm d}(z_{\rm eff} = 1.47) = 24.6 \pm 1.0$.

Figures (13)

• Figure 1: Redshift distributions of the quasar (blue) and the ELG catalogs (orange) used in this work. Only the ELGs with $z > 1.2$ will contribute to the cross-correlation and those redshifts are highlighted by the hatched histogram.
• Figure 2: A smoothed quasar spectrum from the DESI DR2 dataset at $z=3.26$ (TargetID = 39627334130598165). The spectrum shows the emission peaks and forest regions for the two areas typically used in BAO analyses: Ly$\alpha$ (green) and Ly$\beta$ (yellow). In this analysis, we measure C iv absorption in the C iv region (purple) from 1420 to 1520 Å in the quasar rest-frame, and in the Si iv region (blue) from 1260 to 1375 Å.
• Figure 3: Two-dimensional correlation function shown as $r_\parallel$ and $r_\perp$ for the two correlations: CIV(SB2)$\times$QSO and CIV(SB1)$\times$QSO. The BAO feature is not easily seen by eye but is observed as a half ring at $r \sim 100$$h^{-1}$ Mpc. Both correlations are multiplied by $r$ for visualization purposes. There are oscillatory features at $r_\parallel < 0$ in CIV(SB2) caused by quasar redshift errors (discussed in section \ref{['subsec:fit']}).
• Figure 4: Two-dimensional correlation function shown as $r_\parallel$ and $r_\perp$ for the two correlations: CIV(SB2)$\times$ELG and CIV(SB1)$\times$ELG. The BAO feature is not easily seen by eye but is observed as a half ring at $r \sim 100$$h^{-1}$ Mpc. Both correlations are multiplied by $r$ for visualization purposes. The data is noisy so it is difficult to see any oscillatory features at $r_\parallel < 0$ in CIV(SB2).
• Figure 5: Weighted one-dimensional correlation averaged over bins of total separation $r$ and lines of sight $\mu$ when using quasars (blue) and ELGs (orange) as tracers. For each tracer, we are showing the correlation when combining the two side-bands, and we compress the two-dimensional data and best-fit model into one wedge between $-0.95 \leq \mu \leq 1.0$. We multiply both correlations by $r^2$ for visualization purposes. The BAO "peak" is seen as a trough near 100 $h^{-1}$ Mpc for both tracers.