Void Galaxies and AGN Activity in ZOBOV-identified TNG300 Voids Out to z=3.0

Abstract

We study void galaxies in the TNG300 simulation between redshifts $z=3$ and $z=0$. Cosmic void catalogs were constructed using a watershed-based void-finding algorithm, and we define four populations of field galaxies for our investigation: [1] galaxies that are members of a watershed void, [2] galaxies that are located within a radius $r \leq 0.8 R_{\rm eff}$ of the center of a void, [3] galaxies interior to spheres centered on void centers that have underdensity contrasts $<-0.8$, and [4] non-void galaxies. We show that population statistics on void galaxy properties can be recovered from watershed-based void catalogs. Differences between galaxy populations are most pronounced interior to the shell-crossing surface (i.e., population [3]) where densities are intermediate to high. Compared to non-void galaxies at all redshifts, the density controlled galaxies are bluer, smaller, more actively star forming, more massive, and less metal enriched. At redshifts $\geq 1$, these differences are less apparent, likely caused by resolution and selection effects incurred by attempting to define a density-controlled sample from a watershed-based void finding algorithm. Further, we investigate the fraction of galaxies with Active Galactic Nuclei (AGN) and find that our density controlled population has AGN fractions that are significantly higher than those of non-void galaxy population ($79.8 \pm 0.4$\% higher at $z=0.0$ and $61.5\pm 0.7$\% higher at larger redshifts).

Figures (13)

• Figure 1: Median number of gas cells and stellar particles per galaxy as a function of redshift. Each panel shows galaxies from different snapshots. Results are shown for all void galaxies (solid purple), non-void galaxies (circles), radius-controlled galaxies (dashed green), and density-controlled galaxies (dash dotted orange). The vertical blue lines indicate the minimum galaxy radii cuts (see text). For visibility, the number of stellar particles have been shifted down by an order of magnitude.
• Figure 2: Histograms showing average tracer underdensities for all voids (solid purple), radius-controlled underdensity regions (dashed green), and density-controlled underdensity regions (dash dotted orange).
• Figure 3: Empirical cumulative $(g-r)$ probability distribution functions for all void galaxies (purple), non-void galaxies (circles), radius-controlled galaxies (green), and density-controlled (orange). Shaded regions show the 1$\sigma$ spread in the data.
• Figure 4: Top: Median $(g-r)$ color for all void (diamonds), non-void (circles), radius-controlled (crosses), and density-controlled galaxies (triangles) as a function of redshift. Bottom: Ratios of the corresponding points for the non-void and void (squares), radius-controlled (triangles), and density-controlled (circles) galaxies.
• Figure 5: Number of galaxies as a function of radius for non-void (circles), all void (purple), radius-controlled (green), and density-controlled galaxies (orange).