DESI Spectroscopy of HETDEX Emission-line Candidates I: Line Discrimination Validation

TL;DR

This work validates the HETDEX line classification for Ly$\alpha$ emitters by cross-checking DESI follow-up spectroscopy of 1157 sources (including 982 LAEs) and demonstrates that the ELiXer framework can achieve interloper contamination below $<2\%$ while recovering about $96\%$ of true Ly$\alpha$ emitters with correct redshifts. The analysis employs consistent Gaussian line fitting across DESI and HETDEX data, assesses line-centroid, flux, and width agreement, and uses DESI redshifts to quantify the accuracy of HETDEX classifications. The results show strong concordance in line properties, precise redshift alignment ($\langle \Delta z/(1+z)\rangle \sim 6.9\times10^{-5}$; RMS $\sim3.3\times10^{-4}$), and highlight practical limitations from aperture effects and centering uncertainties. Overall, DESI validation supports the robustness of HETDEX's untargeted LAE approach and demonstrates the value of cross-instrument spectroscopic checks for line-discrimination in large-scale structure studies.

Abstract

The Hobby-Eberly Dark Energy Experiment (HETDEX) is an untargeted spectroscopic galaxy survey that uses Ly$α$ emitting galaxies (LAEs) as tracers of 1.9 < z < 3.5 large scale structure. Most detections consist of a single emission line, whose identity is inferred via a Bayesian analysis of ancillary data. To determine the accuracy of these line identifications, HETDEX detections were observed with the Dark Energy Spectroscopic Instrument (DESI). In two DESI pointings, high confidence spectroscopic redshifts are obtained for 1157 sources, including 982 LAEs. The DESI spectra are used to evaluate the accuracy of the HETDEX object classifications, and tune the methodology to achieve the HETDEX science requirement of $\lesssim 2\%$ contamination of the LAE sample by low-redshift emission-line galaxies, while still assigning $96\%$ of the true Ly$α$ emission sample with the correct spectroscopic redshift. We compare emission line measurements between the two experiments assuming a simple Gaussian line fitting model. Fitted values for the central wavelength of the emission line, the measured line flux and line widths are consistent between the surveys within uncertainties. Derived spectroscopic redshifts, from the two classification pipelines, when both agree as an LAE classification, are consistent to within $\langle Δz / (1 + z) \rangle = 6.9\times 10^{-5}$ with an rms scatter of $3.3\times 10^{-4}$.

Figures (10)

• Figure 1: HETDEX main (a.k.a "spring") field outlined in blue, with completion status as indicated by orange tiles as of January 2021 when the target list for the sample was selected (the grey tiles were planned but yet to be observed). The two circles show the position of the two DESI tiles chosen for the follow-up.
• Figure 2: Parent and assigned target distributions for the two pointings. The numbers 80869 and 80870 refer to the DESI tileID, which is unique for the combination of pointing and fiber assignments. Gaps in the assigned target distribution where there were available targets in the parent sample are due to fibers being assigned to other secondary target programs or small regions of the focal plane with non-working fiber positioning robots during that period.
• Figure 3: Example DESI spectra of a faint [O$\;$] galaxy at $z=0.318$ (top) and an LAE at $z=2.984$ (bottom) confirmed by visual inspection.
• Figure 4: The top panel shows the distribution of emission-line detection $S/N$ for HETDEX targets. The grey histogram indicates the range in the input target sample, while the blue shows the recovered galaxies with VI$\_$quality$\ge3$. The cumulative recovery fraction as a function of signal-to-noise ratio is shown in dashed grey for the full sample and in dashed black line where some cuts are applied. Only well positioned sources with emission lines $\lambda>3800$ are included in this sample and those that have not been excluded from the sample in later HETDEX catalogs (ie. we only include those emission lines with DEX_FLAG==1). Sources at low $S/N$ are missed due to astrometric uncertainties for low $S/N$ sources and possible false positives. The bottom panels presents the distribution in $g_\mathrm{HETDEX}$ magnitudes of the input and recovered sample, in grey and blue respectively.
• Figure 5: The results of our comparison of ELiXer classifications of HETDEX spectra with the high confidence visually vetted DESI spectra. The $y$-axis gives the percent contamination of the HETDEX LAE sample by misidentified [O$\;$] galaxies, while the percentage of LAEs recovered by the classifier is on the $x$-axis. The labeled points show the corresponding P(Ly$\alpha$) threshold used to configure the ELiXer classifier. The red horizontal dashed line represents the requirements of the HETDEX survey HETDEXsurvey, though higher contamination rates may be acceptable with the use of additional decontamination methodologies Farrow+2021. Despite differences in the redshift and magnitude distributions with the Spectrocopic-$z$ Assessment Sample (SzAS) in davis2023, the curve is very similar.