The role of magnetic fields in ram pressure stripping of satellite galaxies in the circumgalactic medium around massive galaxies

TL;DR

This study addresses how magnetic fields influence ram pressure stripping of satellite galaxies in the CGM of massive hosts by comparing magnetized and non-magnetized cosmological zoom-in simulations of three haloes with $M_{\rm{200c}} \sim 10^{12.5-13}$ M$_\odot$. Using magnetohydrodynamic modeling in AREPO with tracer particles and a uniform seed field $B_0 = 10^{-14}$ G, the authors quantify gas retention, tail evolution, and mixing for satellites on first infall, including a higher-resolution CGM run for one halo. They find no population-wide difference in retained gas but show that the two most massive satellites lose significantly more gas in the absence of magnetic fields, and that magnetic draping suppresses mixing and inhibits condensation in the stripped tails. Magnetic fields also shape the magnetic-field structure around satellites, aligning with tails and leading-edge draping, suggesting that magnetic effects are essential for realistic modeling of ram pressure stripping and metal distribution in the CGM of massive galaxies.

Abstract

The presence of magnetic fields in galaxies and their haloes could have important consequences for satellite galaxies moving through the magnetised circumgalactic medium (CGM) of their host. We therefore study the effect of magnetic fields on ram pressure stripping of satellites in the CGM of massive galaxies. We use cosmological `zoom-in' simulations of three massive galaxy haloes ($M_{\rm{200c}} = 10^{12.5-13}$ M$_\odot$), each simulated with and without magnetic fields. Across our full sample of satellite galaxies (11 with magnetic fields and 10 without), we find that the fraction of gas retained after infall through the CGM shows no population-wide impact of magnetic fields. However, for the most massive satellites, we find that twice as much gas is stripped without magnetic fields. The evolution of a galaxy's stripped tail is also strongly affected. Magnetic fields reduce turbulent mixing, inhibiting the dispersion of metals into the host CGM. This suppressed mixing greatly reduces condensation from the CGM onto the stripped tail. By studying the magnetic field structure, we find evidence of magnetic draping and attribute differences in the stripping rate to the draping layer. Differences in CGM condensation are attributed to magnetic field lines aligned with the tail suppressing turbulent mixing. We simulate one halo with enhanced resolution in the CGM and show these results are converged with resolution, though the structure of the cool gas in the tail is not. Our results show that magnetic fields can play an important role in ram pressure stripping in galaxy haloes and should be included in simulations of galaxy formation.

Figures (9)

• Figure 1: Edge-on projections of the hydrogen number density (top) and mass-weighted temperature (bottom) for 3 different galaxy haloes at $z=0$. We assume solar metallicity, $Z_\odot = 0.0127$. All haloes are rotated so the central galaxy is displayed 'edge-on'. All haloes are projected at a depth of 200 kpc. Haloes 7 and 8 have a halo mass $M_{200c} \approx 10^{13}$ M$_\odot$, while Halo 3 has a halo mass of approximately $10^{12.5}$ M$_\odot$. See Table \ref{['tab:halo_table']} for full details. The simulations in the top row of each group of panels include magnetic fields, whereas those in the the bottom row do not. Halo 8 was resimulated with 1 kpc spatial refinement in addition to the existing mass-refinement target. The virial radius, $R_{\rm{200c}}$, for each halo is indicated by a dashed white circle. Haloes without magnetic fields generally have higher gas densities in the CGM, particularly in the vicinity of the central galaxy. Temperatures are slightly higher in outer regions with magnetic fields but are otherwise similar. Given that Halo 3 has a lower halo mass, its virial radius and CGM temperature are lower than for the more massive haloes. Additionally, the central galaxy gas discs are smaller without magnetic fields and satellite galaxy tails appear denser and broader.
• Figure 2: Same as Fig. \ref{['fig:all_haloes_nh-t']} but for metallicity (top) and magnetic field strength (bottom), both mass-weighted. Metals are more smoothly distributed in simulations without magnetic fields. The magnetic field strengths are similar in all haloes. Higher magnetic field strengths are found in the central and satellite galaxies. The cool gas in satellite tails has lower field strengths than the gas in galaxies, but higher than the ambient hot halo gas. In Halo 3, we see a region of high magnetic field strength extending from the galaxy to the virial radius. This is co-spatial with cool, metal-rich gas expelled from the galaxy disc. Therefore, it is clear that not all regions of higher magnetic field strength in the CGM are due to ram pressure stripping of satellites.
• Figure 3: The top panel shows the gas mass evolution of Halo 8: Subhalo 1, with and without magnetic fields, as a function of time since crossing $R_{\rm{200c}}$ of the main halo. It takes approximately 1 Gyr for the satellite to travel between $R_{\rm{200c}}$ and its (approximate) pericentric distance, $R_{\rm{min}}$, where $R$ is the distance between the satellite and the central galaxy. With $B$ fields, ${\sim}20\%$ of total gas mass (ISM+CGM) is lost in this time. Without $B$ fields, ${>}50\%$ of gas is stripped. The dotted grey line shows the ram pressure experienced by the subhalo in the simulation with magnetic fields. The ram pressure is calculated as the product of the mean density ahead of the subhalo, and the square of the subhalo velocity. The evolution of the ram pressure is similar without magnetic fields. The middle panel shows that the star-forming gas mass remains reasonably stable in both simulations indicating that most gas is lost from the satellite's CGM, not the ISM. The bottom panel shows the retained gas fraction at pericentre as a function of the distance between pericentre and the central galaxy for a sample of satellite galaxies with total pre-stripping mass, $M_{\rm{total, init}} \ge 3 \times 10^{10}$ M$_\odot$, and which meet the other criteria given in Sec. \ref{['sec:methods']}. As with Halo 8: Subhalo 1, we follow each satellite from $R_{\rm{200c}}$ to pericentre. As expected, galaxies with smaller pericentric distances are more strongly stripped. Median values suggest stripping is slightly stronger without $B$ fields. However, we see considerable scatter and the two populations are consistent with each other overall.
• Figure 4: Projections of the hydrogen number density, temperature, metallicity and vorticity for the largest satellite of Halo 8 with (top panel) and without (bottom panel) magnetic fields. We project this satellite with a projection depth of 50 kpc. Vorticity field lines are overlaid using the Line Integral Convolution method cabral1993. The galaxy is rotated such that the satellite velocity vector points in the negative $z$-direction. The satellites are shown when they have reached pericentre and thus the moment of maximal stripping for the satellites. The stripped tail is much more extensive and the satellite galaxy disc is smaller in the simulation without magnetic fields. The metallicity and, to a lesser extent, the vorticity, show elevated values in the CGM of the host halo around but outside the tail in the simulation without magnetic fields. This indicates increased turbulence between the ram pressure stripped gas and the host CGM. Dashed white rectangles denote the region from which tracer particles are selected in Sec. \ref{['sec:tracer-analysis-results']}.
• Figure 5: Fractions of tracer particles in the tails of two satellite galaxies (Halo 8: Subhalo 1 and Halo 7: Subhalo 1) stripped from each satellite galaxy and its CGM (blue bars) or condensed from the CGM of the host halo (orange bars). Fractions are displayed for simulations with (1st and 3rd sets of bars) and without (2nd and 4th sets) magnetic fields. In the simulations without magnetic fields, the gas condensed from the host CGM makes up a larger fraction of the gas in the tails than in the simulations with magnetic fields. This shows that mixing is more efficient in the absence of magnetic fields.