SIRIUS: Dark matter cusp evolution in dense dwarf galaxies

TL;DR

This study investigates how baryonic physics shapes the inner dark matter density profiles of dwarfs with $M_{\rm vir} \sim 10^9\,M_{\odot}$ using cosmological zoom-in simulations at unprecedented resolution ($m_{\rm gas}=2.37\,M_{\odot}$, $m_{\rm DM}=12.8\,M_{\odot}$). By comparing DM-only and hydro runs for two halos (Halo 230 and Halo 284), the authors find that baryons can either leave the cusp largely intact or deepen it, with Halo 230 showing a modest cusp and Halo 284 developing a much steeper cusp ($\alpha_{\rm Hydro} \approx -1.94$). Neither model exhibits a cusp-to-core transition within the simulated epoch; early halo growth and the timing of star formation appear crucial in determining the central DM response. A higher-resolution run confirms cusp-slope convergence and highlights how early baryonic assembly can yield ultra-compact stellar systems, bridging observations of MW satellites and UCD-like dwarfs. These results imply a strong dependence of inner DM structure on formation history in the $10^9\,M_{\odot}$ regime and motivate further high-resolution studies across a broader halo ensemble.

Abstract

Dwarf galaxies have a wide variety of structures, such as dark matter (DM) distribution, stellar-to-halo mass ratio, and stellar density. Recent high-resolution simulations have shown a variety of stellar-to-halo mass ratios for dwarf galaxies with a DM halo mass of $\sim 10^9 M_{\odot}$ at $z=0$. In this study, we performed cosmological $N$-body/smoothed-particle hydrodynamic zoom-in simulations of dwarf galaxies with the highest gas and DM particle mass resolutions of 2.37 $M_{\odot}$ and 12.8 $M_{\odot}$, respectively. The stellar-to-DM halo mass ratio of one of our simulated dwarf galaxies was $\sim 10^{-4}$, typical for satellites of the Milky Way. The stellar mass ($10^5 M_{\odot}$) and half-mass radius (68 pc) were also similar to those of the satellites of the Milky Way. The power-law slope of the DM halo was $α= -1.1$. On the other hand, the other simulated galaxy exhibited a stellar-to-halo mass ratio of $\sim 10^{-3}$ and a steeper power-law slope ($α=-1.9$) than the other; the presence of baryonic matter deepened the cusp. The mass of $>10^6 M_{\odot}$ and a half-mass radius of $\sim 36$ pc of this galaxy were similar to those of ultra-compact dwarf galaxies rather than the satellites of the Milky Way. This DM halo grew in mass earlier than the former one, and the central DM density was higher than that of the other even in the DM-only simulations.

Figures (16)

• Figure 1: Top panels: Dark matter surface density distribution, and the positions of Halo 230 and Halo 284 in it. The red crosses show the centers of the halos at $z=0$. Bottom panels: Zoomed-in images of each halo. The white circles show the virial radii. Axis scales are in $\mathrm{Mpc} / h$ for all panels. Alt text: Four-by-four panel figure. The top panels show the positions of Halo 230 and 284 in a four comoving megaparsec square scale large-scale distribution of dark matter surface density in x-y (left) and x-z (right) planes. Bottom panels show zoomed-in images of Halo 230 (left) and Halo 284 (right).
• Figure 2: DM surface density distributions of Halo 230 (left) and Halo 284 (right) at $z=0.5$. From top to bottom, each row represents the results of the DMO, Hydro, and Hydro Low models, respectively. The left and right columns of each halo show the DM surface density within $0.5\,R_{\mathrm{vir}}(t)$ and $5\,R_{\mathrm{vir}}(t)$ from the halo center, respectively. The red dots in the left column indicate star particles. Axis scales are in $\mathrm{Mpc} / h$ for all panels. Alt text: Twelve-panel grayscale images of the dark matter surface density centered on the halo center. Red dots are plotted on the grayscale images for the Hydro and Hydro Low models.
• Figure 3: Time evolution of the viral mass of the halo ($M_{\mathrm{vir}}$) for Halo 230 (left panel) and Halo 284 (tight panel). The green dotted vertical line indicates the reionization epoch assumed in this simulation ($z=8.5\,$, $t=0.588\,\mathrm{Gyr}$). Alt text: Two line graphs. The range of the x-axis (time) is zero to fourteen gigayears. Blue and red lines are for DMO and Hydro runs, respectively. The virial masses increase with time and reach the final mass at around eight point five gigayears.
• Figure 4: Star formation histories of Halo 230 (left) and Halo 284 (right). Star formation rates are averaged over $50\,\mathrm{Myr}$. The green dotted line indicates the reionization epoch. Alt text: Line graphs peaked at around the reionization epoch. The red and orange lines correspond to the Hydro and Hydro Low models, respectively. The peak value of the star formation rate of Halo 230 is slightly less than ten to the minus three solar mass per year. The peak value of Halo 284 is slightly less than ten to the minus two solar masses per year.
• Figure 5: Time evolution of $V_{\mathrm{max}}$ and $T_{\mathrm{vir}}$ in each simulation. The left axis shows the maximum circular velocity, whereas the $y$-axis tics on the right side show the virial temperature. The green dotted line represents the time at which reionization occurs in this simulation. Alt text: Line graph with two lines. Purple and green lines are for Halo 230 and 284, respectively. The X-axis indicated the time in gigayears. Left and right y-axes show the circular velocity and the corresponding virial temperature, respectively. The circular velocity increases with time and reaches a circular velocity of approximately 17 kilometers per second. The circular velocity corresponds to ten thousand kelvin.