The LMC Corona as Evidence for a First Passage

TL;DR

This study tests whether the LMC’s circumgalactic medium preserves signatures of its orbital history around the Milky Way by running constrained idealized simulations with fixed MW/LMC potentials along two orbital histories (first passage and second passage) and evolving the gas halos self-consistently. Synthetic absorption sightlines generated with Trident are compared to Mishra et al.’s measurements of the LMC Corona, focusing on velocity and column-density profiles and a truncation radius $ ho_T$. The results show a best-match for the first-passage model with $ ho_T=15.7^{+0.7}_{-1.0}$ kpc (in the Mishra et al. range of $17$–$20$ kpc), while the second-passage model yields $ ho_T=8.5 extpm1.5$ kpc and underpredicts velocities. Overall, the findings favor a first infall scenario for the LMC and have implications for MW–LMC CGM interactions and future observational tests of the Magellanic Corona.

Abstract

We use constrained idealized simulations of the LMC/Milky Way interaction to determine if the size of the LMC's gaseous halo (Corona) can be used to distinguish between first and second passage models $-$ an orbital trajectory for the LMC in which it has just recently approached the Milky Way for the first time (first passage), or one in which it has had a previous pericenter (second passage). Using live circumgalactic gas particles combined with analytic dark matter potentials evolved to follow previously published orbital trajectories, we find that the first passage model is able to reproduce the observed velocity profile and column density profile of the present day LMC Corona. On the other hand, in a second passage scenario the longer interaction time leads to the velocities and column densities around the LMC at the present day being too low. Based on this observed velocity profile, recent works have found that the LMC's Corona has been truncated to 17$-$20 kpc, and we find truncation radii of $15.3\pm 0.9$ kpc and $7.6\pm 2.0$ kpc for the first and second passage models, respectively. Thus, based on the gas properties of the LMC's CGM at the present day, a second passage trajectory is disfavored.

Figures (4)

• Figure 1: Projected gas density in the first and second passage models at the present day. The top and bottom panels show the MW and LMC CGM gas column density in the Cartesian $y-z$ plane, respectively, with the first passage model on the left and the second passage model on the right. The orbital trajectories of the LMC and MW are drawn in white lines while their present-day positions are marked with plus symbols.
• Figure 2: On-sky projection of the present-day Magellanic Corona and its velocity profile. The top panels show the H2 gas column density in Magellanic Coordinates for the first passage model on the left and the second passage model on the right. The white box denotes the region from which the random sightlines were selected to best match the region probed by the observational data in mishra24. The bottom panels show the gas LSR velocity distribution as a function of Magellanic Longitude for the first (left) and second (right) passage models.
• Figure 3: LSR velocities and column densities of mock observations of the simulations compared against the data. The top panels show the column density weighted LSR velocities, and the bottom panels show the H2 column densities, both as a function of impact parameter from the LMC. The left panels show the results for the first passage model (in red), while the right panels are for the second passage model (in orange). The best fit linear regression to the data points is shown as a dashed line with the 95% confidence interval shown as the red/orange shaded region. The blue stars in the top panels are the observational data points from mishra24 for the C4 detections and the blue vertical band is their quoted truncation radius. The grey horizontal band shows the LMC's systemic velocity $\pm50$ km s$^{-1}$ which was used to determine the truncation radius (see text). In the bottom panels, the blue region is the fit from dk22 after using cloudy modeling to extrapolate the H2 densities. For the top panels and both fits, we have only included mock sightlines with H2 column densities above $10^{17.5}$ cm$^{-2}$. The downward arrows in the bottom panels show the $\rho$ values for the sightlines with log $N_\mathrm{HII}<17.5$ which were not used in the analysis. There were 51 and 48 sightlines remaining for the first and second passage models, respectively.
• Figure 4: Distributions of truncation radii ($\rho_T$) for the first and second passage models compared against the range found in mishra24 ($17-20$ kpc, shown in blue). As before, the first passage model is shown in red, and the second passage model is shown in orange. The curves are presented as skewnorm distributions where we have calculated the root semivariances by finding the range of $\rho$ values where the 68% confidence interval of the $v_\mathrm{LSR}$ vs $\rho$ fit crosses the $v_\mathrm{LMC}-50$ km s$^{-1}$$=230$ km s$^{-1}$ line (see top row of Figure \ref{['fig:velfits']}). We find $\rho_T=15.7^{+0.7}_{-1.0}$ kpc for the first passage model, and $\rho_T=8.5\pm1.5$ kpc for the second passage model.