Linking enhanced star formation and quenching to faint tidal features in galaxies

TL;DR

This study quantifies how faint tidal features from galaxy mergers relate to star formation and quenching, using a large, morphology-resolved sample drawn from DECaLS and CFHTLS and cross-matched to SDSS-derived properties. By classifying evolutionary phase with PCA and identifying AGN via BPT diagnostics, the authors show that tidal galaxies are markedly more likely to host starbursts ($ imes$6.6) and post-starbursts ($ imes$19.6) than mass–redshift-matched non-tidal controls, with clear morphology-dependent trends (e.g., shells favor quiescence, arms favor star formation). They find no statistically significant overall AGN excess in tidally disturbed systems, though an alternative normalization including low-SNR cases suggests a potential increase, highlighting sensitivity to methodological choices. Overall, the results reinforce mergers as drivers of both enhanced star formation and rapid quenching, and demonstrate that the strength of these effects depends on the detailed interaction morphology traced by tidal features.

Abstract

Galaxy mergers and interactions have long been suggested as a significant driver of galaxy evolution. However, the exact extent to which mergers enhance star formation and AGN activity has been challenging to establish observationally. In previous work, we visually classified a sample of galaxies with various types of faint tidal features in DECaLS images. In this paper, we cross-correlate this sample with SDSS-derived data to investigate how the presence and specific nature of these features correlates with intense star formation and AGN activity. Averaged over all tidal classes, we find that our 688 tidal feature galaxies are 6.6$\pm$0.9 times more likely to be in a starburst phase and 19.6$\pm$5.0 times more likely to have rapidly quenched (post-starbursts) than a sample of 4073 controls matched in both stellar mass and redshift. Examining differences between tidal classes, galaxies with \textit{arm} features were $\sim$1.3-4.0 times more likely to be starbursting than the other categories, while those with \textit{shell} features were $\sim$2.3-5.3 times more likely to be in a quiescent state. In a similar analysis, we identify which galaxies show evidence of AGN activity (from a sample of $\sim$2100) and find no significant difference between those with or without tidal features. Overall, our results reinforce the notion that mergers play an important role in driving star formation and rapid quenching in galaxies, and provide some of the first empirical evidence that the strength of this effect has a dependence on the detailed nature of the interaction, as traced by the tidal feature morphology.

Figures (8)

• Figure 1: Example thumbnails of the four categories of tidal features used in this work: arms (top), streams (middle-top), shells (middle-bottom), and diffuse (bottom). The images on the left were created by 2022MNRAS.509.3966W and those on the right were created in 2024MNRAS.534.1459G.
• Figure 2: The normalised distribution of Sérsic indices for the DECaLS sample split between those identified as ellipticals and discs (spiral, edge-on, or miscellaneous) by the 2022MNRAS.509.3966W classifier.
• Figure 3: Stellar mass ($M_\star$)-redshift ($z$) distribution of the sample. Points are shaded such that darker points indicate a higher concentration of galaxies. The contours show the region containing 50 (thicker) and 90 (thinner) per cent of the data. The top and side panels show the individual kernel density estimates of stellar mass and redshift. The dashed and dotted lines indicate the distribution for the tidal sample and the non-tidal controls, respectively.
• Figure 4: The evolutionary phase based on the 2007MNRAS.381..543W principal component analysis (PCA). See the text for details of the selection criteria based on the first and second principal components (PC$_1$ and PC$_2$). The galaxies are separated into those identified as post-starbursts (PSB), starbursts (SB), star-forming (SF), those intermediate between SB and SF (SB/SF), those in the green valley (GV), and those that are quiescent (Q). Galaxies that were outside of these classifications were considered as not specified (NS). Where the error on PC$_1$ or PC$_2$ crossed the boundary of a phase, the galaxy was considered as unidentified (U).
• Figure 5: Baldwin–Phillips–Terlevich 1981PASP...93....5B optical emission line analysis for the determination of active galactic nuclei (AGN) activity. The sample was split into galaxies with AGN, star-forming galaxies (SFG), and composites (Comp) using the 2006MNRAS.372..961K criteria.