ODIN: Spectroscopic Validation of Ly$α$-Emitting Galaxy Samples with DESI

Abstract

The One-hundred-deg^2 DECam Imaging in Narrowbands (ODIN) survey is conducting the widest-field deep narrow-band imaging of the equatorial and southern skies. ODIN uses three custom-built narrow-band (NB) filters that sample Lya-emitting galaxies (LAEs) within thin cosmic slices centered at z=2.4, 3.1, and 4.5. In this work, we utilize extensive DESI spectroscopy of ODIN-selected galaxies in the COSMOS and XMM-LSS fields to validate our LAE selection. 2-4 hr exposures with DESI yielded redshift confirmation of 3,075 ODIN LAE candidates with NB magnitudes brighter than 26~mag. Restricting to objects that yield high-confidence redshifts, the confirmation rates are (93, 96, 92)% at z=(2.4, 3.1, 4.5). The primary contaminants consist of active galactic nuclei at the expected Lya redshift range and lower redshifts (C IV, C III]), with the remainder being star-forming galaxies ([O II] and [O III]). We find minimal contamination from [O II] emitters in our sample (<~1%), implying that our REW>20 A narrow-band excess photometry requirement is sufficient to remove them.

Figures (6)

• Figure 1: Maps of two ODIN fields (COSMOS and XMM-LSS) showing the locations of spectroscopically confirmed LAEs. The objects are colored as follows: All SE objects in good regions (gray); Tr-only LAEs in good regions (not starmasked and in regions containing high-quality HSC Broadband data) (orange); and Tr+ODIN LAEs (green). Of the confirmed LAEs not selected by ODIN’s criteria, roughly half lie in sky regions excluded by ODIN (either in starmasks or regions without HSC broad-band data). Counts of the subsets of Tr-only LAEs can be found in \ref{['tab:spec-lae-distribution']}.
• Figure 2: Example spectra showcasing the confirmed LAEs and primary contaminants of our sample (AGN at lower-$z$, AGN at target-$z$, Milky Way white dwarfs (WD), [O$\;$] emitters, and [O$\;$] emitters). Flux (black) and noise (gray) are boxcar smoothed to exemplify the sources' prominent emission line features, which are denoted by dashed gray lines. The noise spectra were turned into variance spectra and given the same boxcar smoothing. Above each spectrum is the associated DESI target ID, and the panel legend lists the NB filter from which the target was selected, the type of object, redshift ($z$), and quality of the spectrum ( q).
• Figure 3: Color-magnitude diagrams for various subsets of objects color-coded by spectral classification determined from the visual inspection campaign (Section \ref{['sec:results']}). The left column shows ODIN LAE candidates using SE photometry. The other columns show DESI-targeted sources that are in the ODIN SE catalog but not in the ODIN LAE sample, with the middle column using SE photometry and the right column using Tr photometry. From top to bottom, we display the field/filter combinations: XMM-LSS/$N419$, COSMOS/$N419$, COSMOS/$N501$, and COSMOS/$N673$. The x-axis shows each object's narrowband (NB) magnitude and the y-axis shows the excess NB flux density in units of magnitudes. Each plot shows confirmed LAEs (red), confirmed AGN in the target redshift range (yellow), confirmed AGN outside the target redshift range (purple), other confirmed emission line galaxies and stars (blue), and targeted objects without identification (orange). The horizontal dashed line in each panel represents a NB excess threshold corresponding to a rest-frame Lyman Alpha equivalent width of 20 Å. The NB excess in the left and middle columns is calculated with a weighted double broadband magnitude ($rg-N419$, $gr-N501$, or $gi-N673$; see firestone24). The NB excess for the (rightmost) Tr photometry column is calculated with a single broadband ($g-N419$, $g-N501$, or $r-N673$; described A. Dey et al. in prep.). We note that the majority of Tr spec-LAEs which were not selected as SE LAE candidates did not have sufficient narrowband excess to satisfy the cut.
• Figure 4: Redshift distribution of DESI-confirmed (${\tt q}\geq 2.5$) ODIN sources observed across all three NB filters. Left: A histogram of the entire redshift range showing contaminants that are not within the Ly$\alpha$ redshift window defined by the transmission curve of the NB filter. The dashed lines indicate the redshifts associated with common non-Ly$\alpha$ emission lines. Right: Zoomed-in redshift histogram of Ly$\alpha$ emitting objects. The black curves represent the transmission functions of the NB filters.
• Figure 5: Histograms displaying Tr+ODIN LAE candidates sorted by narrowband magnitude and colored by classification. Narrowband magnitude is displayed in bins of size 0.2 and ranges from 20 to 26.2. The colors are consistent with those in Figure 3. The left column displays the raw histograms stacked on top of one another. The right column displays the normalized histograms for the same bin sizes. At dimmer magnitudes, a higher percentage of the ODIN LAE Candidates are targeted objects without identification, and at brighter magnitudes, there is a higher percentage of contaminants.