1 Introduction↩︎

As large-scale structure grows, the tidal fields that form these structures induce effects on the shape, spin, and orientation of galaxies. These effects are classified as Intrinsic Alignments (IA) [1], [2]. IA traces cosmological effects on the large-scale tidal field [3], including correlating shapes to the quadrupole of projected large-scale structure. IA also applies to galaxy ensembles, such as galaxy groups, clusters, and halos [4]–[6].

To maximize the number of galaxy ensembles, multiplets can be used. These are smaller than groups and not necessarily virialized. Multiplets can be better for measuring alignment than individual galaxies when samples: have poor imaging, are dense, or have little individual alignment [7]. Multiplet alignment is a projected quantity, correlating multiplets’ orientations in the plane of the sky with the projected separation to tracer galaxies. Multiplets can be constructed without the full 3D positions of each multiplet member, since multiplet orientation is obtained from projected positions. Here, we explore whether multiplets made with imaging and spectroscopic data can produce the same signal as multiplets identified with only spectroscopic data.

2 Data↩︎

DESI constrains structure growth and dark energy by obtaining millions of extragalactic redshifts [8]–[11]. We use two catalogs: the DESI Y1 Luminous Red Galaxy (LRG) Sample [12], which contains spectroscopic redshifts, and the DESI Legacy Imaging Survey [13], [14] which consists of the imaging information used in DESI’s target selection, including \(r\) and \(W1\) photometry [15].

We chose LRGs since they have high bias and display strong multiplet alignment. DESI’s Y1 catalog consists of 8 million LRGs in the redshift range of 0.4 \(< z < \sim\) 1.0. It is the same survey used in the initial detection of multiplet alignment [7], allowing us to make a comparison to the spectroscopic only case. From the imaging catalog, we select LRG-like galaxies that are twice as faint as the galaxies in the spectroscopic catalog by using original cuts for the main survey, but with a lower fiber magnitude cut of \(z<22.35\). Our resulting LRG imaging sample has a density 9.4 times higher than the spectroscopic sample (605 deg\(^{-2}\)).

3 Methods↩︎

We identify imaged galaxies close to spectroscopic targets and measure the alignment of these multiplets relative to positions of the full LRG spectroscopic sample. We explore three limits on the sky distance between imaged and spectroscopic galaxies: 0.6 arcminutes, 1.8 arcminutes, and 3 arcminutes. These correspond to a transverse comoving distance of about 0.3 - 1 Mpc in the LRG redshift range, the distance used to define spectroscopic multiplets. The members of multiplets are further limited by the difference in color compared to the spectroscopic galaxy, increasing the probability that they are at the same redshift. The two color-based cuts we explore are \(\Delta (r-W1)\) less than 0.1 and 0.3. From combinations of these cuts, we create six enhanced multiplet catalogs. The multiplet alignment signal is defined and calculated following [7] which we summarize here. The key difference is that we assign the imaged neighboring galaxies the same redshift as their center galaxy. The projected orientation of multiplets is determined from members’ sky positions, then correlated with the positions of all LRGs in the spectroscopic catalog. The distance along the line of sight between the multiplet’s center and a tracer, \(\Pi_{\rm max}\), increases with their projected separation, \(R\). Alignment is described by \(\cos(2\phi)\), measured as a function of \(R\). \(\phi\) is the projected multiplet orientation relative to tracers.

4 Results↩︎

Our results demonstrate that supplementing a redshift catalog with imaging to find multiplets can produce a signal on par with spectroscopic-only measurements. Figure 1 compares the original multiplet alignment from the spectroscopic catalog to measurements of the enhanced catalogs. The signals have the same scale dependence with comparable signal-to-noise ratios (SNR). Multiplets made with the strictest cuts produce the most similar signal to the spectroscopic case. This ‘purest’ multiplet catalog produces the highest correlations, but a more generous cut will produce more multiplets and a higher SNR. Notably, the least strict cuts produce a higher SNR than the spectroscopic measurement. The main factors affecting the signal amplitude are the number of unassociated photometric interlopers and the multiplet size. The alignment of multiplets made with larger cuts are more likely to be diluted by interlopers, but are also larger, which increases alignment strength. While the alignment amplitude varies between our samples, they display identical scale dependence in the linear regime. This can be directly related to the underlying tidal shear for some applications and the amplitude calibrated or otherwise inferred for others. For a more detailed discussion of applications, see [7].

This work presents a proof-of-concept and initial suggestions for which cuts will optimize future work combining imaging with spectroscopic data. Our technique can be used for galaxy populations besides LRGs, and for denser imaging surveys such as LSST [16]. This has the potential to outperform multiplet measurements made with spectra alone.
Data plotted here can be found at zenodo.org/uploads/17298578.

5 Acknowledgments↩︎

This material is based upon work supported by the U.S.Department of Energy under grants DE-SC0013718 and DE-SC0024787, and the Research Experiences for Undergraduates Program under NSF 23-601.

This work is supported by the U.S. Department of Energy, Office of Science, Office of High-Energy Physics, under Contract No. DE–AC02–05CH11231, and by the National Energy Research Scientific Computing Center. Additional support for DESI was provided by the U.S. National Science Foundation, Division of Astronomical Sciences under Contract No. AST-0950945; the Science and Technology Facilities Council of the United Kingdom; the Gordon and Betty Moore Foundation; the Heising-Simons Foundation; the French Alternative Energies and Atomic Energy Commission; the National Council of Humanities, Science and Technology of Mexico; the Ministry of Science, Innovation and Universities of Spain, and by the DESI Member Institutions.

The authors are honored to conduct research on I’oligam Du’ag, a mountain particularly significant to the Tohono O’odham Nation.

For more information, visit desi.lbl.gov.

References↩︎