2022ApJ...926..107M

The Milky Way halo was predominantly formed by the merging of numerous progenitor galaxies. However, our knowledge of this process is still incomplete, especially in regard to the total number of mergers, their global dynamical properties and their contribution to the stellar population of the Galactic halo. Here, we uncover the Milky Way mergers by detecting groupings of globular clusters, stellar streams, and satellite galaxies in action ( J ) space. While actions fully characterize the orbits, we additionally use the redundant information on their energy (E) to enhance the contrast between the groupings. For this endeavor, we use Gaia EDR3-based measurements of 170 globular clusters, 41 streams, and 46 satellites to derive their J and E. To detect groups, we use the ENLINK software, coupled with a statistical procedure that accounts for the observed phase-space uncertainties of these objects. We detect a total of N = 6 groups, including the previously known mergers Sagittarius, Cetus, Gaia-Sausage/Enceladus, LMS-1/Wukong, Arjuna/Sequoia/I'itoi, and one new merger that we call Pontus. All of these mergers, together, comprise 62 objects (≃25% of our sample). We discuss their members, orbital properties, and metallicity distributions. We find that the three most-metal-poor streams of our galaxy-"C-19" ([Fe/H] = -3.4 dex), "Sylgr" ([Fe/H] = -2.9 dex), and "Phoenix" ([Fe/H] = -2.7 dex)-are associated with LMS-1/Wukong, showing it to be the most-metal-poor merger. The global dynamical atlas of Milky Way mergers that we present here provides a present-day reference for galaxy formation models.