Evidence for Intrinsic Galaxy Alignments in Ellipticity Autocorrelations out to $100 h^{-1}\textrm{Mpc}$ from SDSS Galaxies with DESI Imaging

TL;DR

This paper presents the first observational detection of large-scale intrinsic galaxy alignments through ellipticity autocorrelations extending to 100 Mpc/h, providing new insights into galaxy formation and cosmology.

Contribution

It introduces a novel method to measure intrinsic alignments from ellipticity autocorrelations over large scales, enhancing detection significance and opening new observational avenues.

Findings

Detected intrinsic alignments up to 100 Mpc/h

Expanded II correlation function isolates line-of-sight effects

Joint analysis of II(+), II(-) increases detection significance

Abstract

Measuring the autocorrelation of galaxy shapes, known as the intrinsic-intrinsic (II) correlation, is important for both cosmology and understanding the formation of massive elliptical galaxies. However, such measurements are significantly more challenging than those of the cross-correlation with galaxy density (GI correlation) due to the much lower signal-to-noise ratio. In this Letter, we report the first observational evidence for large-scale intrinsic alignments measured from the ellipticity autocorrelations, extending out to $100\,h^{-1}\,{\rm Mpc}$. From the Sloan Digital Sky Survey (SDSS) and SDSS-III Baryon Oscillation Spectroscopic Survey, we analyze, over the redshift range $0.16\leq z\leq 0.70$, luminous red galaxy, LOWZ, and CMASS galaxy samples, the latter two of which are crossmatched with high-quality Dark Energy Spectrograph Instrument imaging data. By expanding one of the two II correlation functions, II($-$), in terms of the associated Legendre polynomials, we effectively isolate the line-of-sight projection effects and enhance the signal. The resulting correlation for all three samples exhibits a clear power-law form. We also show that jointly analyzing the two II correlations, II($+$) and II($-$), increases the detection significance by $\sim 10\%$, even though both are derived from the same $E$-mode power spectrum. Importantly, this measurement opens a new observational window for probing signals uniquely encoded in shape autocorrelations, such as tensor perturbations from the gravitational waves. Our analysis establishes a practical framework for extracting such effects.