Caught in the web: galaxy mergers along cosmic filaments

Abstract

Galaxy clusters grow through the accretion of galaxies from groups, filaments, and other clusters. During this process, galaxies may undergo pre-processing in lower-density environments, where galaxy-galaxy mergers and other interactions can significantly alter their properties prior to cluster infall. We investigate the role of galaxy mergers in the pre-processing of galaxies prior to cluster infall by studying the spatial distribution of mergers across the cosmic web. We use a sample of 43,922 galaxies targeted by the 4MOST CHANCES survey in and around 33 low-redshift clusters (z < 0.07). Using Zoobot, a deep-learning framework trained on Galaxy Zoo data, we identify 698 galaxy mergers. We measure their distances to cosmic web filaments and compare them with those of non-merging galaxies. We find that galaxy mergers are significantly closer to filaments than the non-merging galaxy population, with this trend being strongest beyond the cluster virial radius. This suggests that filaments provide conditions conducive to mergers, possibly moderating relative velocities and enhancing gas availability. Our findings support a scenario in which filaments play a key role in transforming galaxies through pre-processing by promoting mergers before they enter cluster cores where star formation quenches.

Figures (7)

• Figure 1: Normalized distribution of $D_{fil}$ for all merging galaxies (orange dashed line), compared to the non-merging galaxy population (blue dotted line). Left: Distribution for galaxies across the entire cluster region (r < $5\times R_{200}$); Middle: galaxies in the inner cluster region (r < $R_{200}$). Right: galaxies in the cluster outskirts ($R_{200}$ < r < $5\times R_{200}$). For visualization, distances to filaments are truncated at $2\,\mathrm{Mpc}$, beyond which the galaxy density drops consistently below $\sim 0.2$. Error bars correspond to Poisson uncertainties in each bin, while vertical lines mark the median of each distribution. Shaded regions indicate the 68% bootstrap confidence intervals (percentiles 16–84) around the median. The AD statistics and p-values comparing both distributions are shown in the legend. The AD statistics show that galaxy mergers are preferentially located closer to filaments than non-merging galaxies, both at all radii and specifically outside the virial region. Distances to filaments are truncated at 2 Mpc in the plots for visibility, although the full distributions extend up to $\sim$4 Mpc.
• Figure 2: Distribution of galaxy mergers in the example cluster Abell 85. Top: Spatial distribution of galaxies in the cluster out to $5 R_{200}$ showing merging galaxies (orange stars) and non-merging galaxies (blue dots) relative to intracluster filaments (black lines). The pink circle denotes the $R_{200}$ radius for reference. Bottom: Normalized distribution of $D_{fil}$, comparing merging galaxies (orange) with the non-merging galaxy population (blue) at r < $5\times R_{200}$. Symbols are as in Figure \ref{['fig:all']}. In this cluster mergers are significantly closer to filaments by a factor of 2.6 compared with non-merging galaxies.
• Figure 3: Schematic representation of the calculation of the minimum distance ($D_{fil}$) between a galaxy (blue) and a filament (black dashed curve). The filament is modelled as a set of linear segments (pink line between pink points). The shortest perpendicular distance is computed using the point line method, and is taken as the galaxy--filament distance.
• Figure 4: Examples of randomly selected merging galaxies. The upper panel shows galaxies selected by $F_{\mathrm{MM}}$, and the bottom panel shows galaxies selected by $F_{\mathrm{MD}}$. In both panels, galaxies are arranged from left to right in decreasing order of the corresponding score.
• Figure 5: Stellar mass distributions of merging and non-merging galaxies. The top panel shows the original samples, which are similar in shape but statistically different (AD and KS tests). The bottom panel shows a ratio-matched control sample, where non-mergers are randomly down-sampled in stellar-mass bins to reproduce the merger mass distribution without enforcing equal sample sizes. The result corresponds to one realization from 1000 Monte Carlo resamplings. Controlling for stellar mass does not change the conclusions.