Large Scale Structure and Environmental Effects on Dwarf Galaxy Growth

Abstract

Dwarf galaxies serve as key models for understanding galaxy assembly in the early universe, with their final properties influenced by environmental factors. Using the dark matter-only simulation "Copernicus Complexio" (COCO) and the semi-analytic model GALFORM, we examine the stellar mass assembly of dwarf galaxies across different cosmic web regions, defined by the NEXUS+/CACTUS algorithm. We identify significant variations in stellar mass assembly based on final mass, with the largest dwarf galaxies assembling, on average, 50% of their mass 7.7 Gyrs later than the smallest ones. Central galaxies also differ in their assembly from satellites of comparable final mass, forming 50% of their mass 2.5 Gyrs later. The location within the cosmic web further influences assembly, with satellite galaxies showing greater differences than centrals. Satellites in the densest regions assemble their mass 1.5 Gyrs earlier than those in the least dense regions, compared to 0.69 Gyrs for central galaxies. This disparity arises from varying infall times, with satellites in dense environments infalling 5.2 Gyrs earlier than those in voids. Additionally, we investigate the impact of reionisation parameters, specifically the timing ($z_{cut}$) and filtering scale ($v_{cut}$) of reionisation. The stellar-to-halo-mass relation shows a power law break between $10^8~\mathrm{M}_\odot < M_{200} < 10^{10}~\mathrm{M}_\odot$, with earlier $z_{cut}$ or higher $v_{cut}$ leading to more star formation suppression in lower-mass haloes. The halo occupation fraction is also affected, with later $z_{cut}$ or lower $v_{cut}$ resulting in fewer lower-mass haloes being occupied at $z=0$. Our investigation provides a valuable theoretical framework for interpreting upcoming observational data in this mass regime.

Figures (18)

• Figure 1: The luminosity function at $z=0$ for all galaxies in the fiducial model matched with the $b_J$-band and $K$-band luminosity functions from norberg2002 and driver2012, respectively. Parameters in GALFORM are tuned to match these luminosity functions by adjusting a few parameters in the model by a small amount, in the regime where data is available, detailed in lacey2016.
• Figure 2: The stellar mass function of all galaxies in the COCO/GALFORM fiducial model, shown compared with the EAGLE simulation schaye_2015, and observational data compiled by behroozi2019, which includes data from moustakas2013 and baldry2012. There is a good agreement between GALFORM and EAGLE, and reasonable consistency with the observational data. Note, however, a detailed comparison with the inferred stellar mass function at $z=0$ is not straightforward. This is because GALFORM uses two different initial mass functions (IMFs), one for starbursts and one for quiescent star formation, whereas a single IMF is typically assumed in observations. The nature of this SMF is bimodal, with the two populations being the reionisation relics and the larger galaxies. See Section \ref{['sec:mass_ass']} for more information on these groups. The dip in the SMF corresponds to the mass scale where reionisation affects galaxy formation in the model. The labels XS, S, M and L correspond to the mass bins used later in Section \ref{['sec:fiducial_results']}.
• Figure 3: The $z=0$ stellar-to-halo-mass relation for central galaxies in the fiducial model used in this paper, compared to the relations from fattahi2017 and behroozi2019. The blue line is the 50th percentile of the COCO/GALFORM data, with the shaded region representing the 16th to 84th percentiles. The relation is very steep from $10^{4} - 10^{6}$M$_{\odot}$, meaning that galaxies with vast differences in stellar mass can be formed within haloes of very similar final masses, as is also seen in Figure \ref{['fig:RI_halogrowth']}. Both the relations from behroozi2019 and from fattahi2017 agree well with the predictions from COCO/GALFORM, showing reasonable agreement between semi-analytics, empirical, and hydrodynamical simulations.
• Figure 4: The COCO volume with different colours denoting the different regions of the cosmic web, as defined by NEXUS+/CACTUS. This is a position plot in the x-y plane, and all points have $z$ coordinates between 30 and 40. The size of the points denoting each galaxy are proportional to the stellar mass of the galaxy, using a log scale. The cosmic web structure is clearly visible.
• Figure 5: The in-situ stellar mass assembly of central dwarf galaxies in COCO/GALFORM, here split into four different stellar mass bins with a log y-axis. The points show the $z_{50}/m_{50}$ and $z_{90}/m_{90}$ points, marking the points at which the stellar mass of the galaxy is at 50% and 90% of its final mass respectively, and the shaded regions represent the 16th to the 84th percentile of assembly at a given redshift. The final day mass of an object has a significant impact on the stellar mass assembly, with the smallest galaxies assembling the majority of their stellar mass before $z=6$, and the largest objects assembling most of their stellar mass after $z=2$. The "S" mass bin are "turnover dwarfs", with masses between lower-mass reionisation relics and "normal" larger galaxies. The largest mass bin, "L" assembles 50% of its mass at a redshift ($z_{50}$) 7.7 Gyrs later (on average) than the smallest mass bin, "XS". The numbers next to each mass bin in the legend represent the number of galaxies in each mass bin.