Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 May;605(7910):435-439.
doi: 10.1038/s41586-022-04665-6. Epub 2022 May 18.

A trail of dark-matter-free galaxies from a bullet-dwarf collision

Affiliations

A trail of dark-matter-free galaxies from a bullet-dwarf collision

Pieter van Dokkum et al. Nature. 2022 May.

Abstract

The ultra-diffuse galaxies DF2 and DF4 in the NGC 1052 group share several unusual properties: they both have large sizes1, rich populations of overluminous and large globular clusters2-6, and very low velocity dispersions that indicate little or no dark matter7-10. It has been suggested that these galaxies were formed in the aftermath of high-velocity collisions of gas-rich galaxies11-13, events that resemble the collision that created the bullet cluster14 but on much smaller scales. The gas separates from the dark matter in the collision and subsequent star formation leads to the formation of one or more dark-matter-free galaxies12. Here we show that the present-day line-of-sight distances and radial velocities of DF2 and DF4 are consistent with their joint formation in the aftermath of a single bullet-dwarf collision, around eight billion years ago. Moreover, we find that DF2 and DF4 are part of an apparent linear substructure of seven to eleven large, low-luminosity objects. We propose that these all originated in the same event, forming a trail of dark-matter-free galaxies that is roughly more than two megaparsecs long and angled 7° ± 2° from the line of sight. We also tentatively identify the highly dark-matter-dominated remnants of the two progenitor galaxies that are expected11 at the leading edges of the trail.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Geometry of the DF2 and DF4 system.
a, The radial velocity difference between DF2 and DF4 is 358 km s−1 and this large velocity difference is accompanied by a large line-of-sight separation of 2.1 ± 0.5 Mpc (refs. ,). The geometry implies that the galaxies are moving away from each other. Tracing their positions back in time, we infer that they were formed in a high-speed encounter ≥6 Gyr ago. b, Example of a collisional scenario involving NGC 1052. Velocities are given with respect to that galaxy (cz = 1,488 km s−1). An infalling gas-rich galaxy on an unbound orbit collided with a satellite of NGC 1052 about 8 Gyr ago, leading to two dark remnants (possibly RCP 32 and DF7), DF2 and DF4, and three to seven other dark-matter-free galaxies.
Fig. 2
Fig. 2. A linear feature in the spatial distribution of faint galaxies in the NGC 1052 field.
a, Distribution of galaxies with g > 16.5 (black circles) from a recent compilation of low-surface-brightness objects in the NGC 1052 field. The positions are with respect to the coordinates of NGC 1052, the central bright elliptical galaxy in the group. b, Hough transform of the spatial distribution. The peak corresponds to a line that has 11 galaxies located within ±30 kpc. The significance of the feature, as determined from randomized realizations of the data, is 97%. c, Zoom of a, shown with the four brightest members of the NGC 1052 group. Yellow circles correspond to the positions of the galaxies in the box shown in a. The orientation and offset of the line corresponds to the location of the peak of the Hough transform. Both DF2 and DF4 are part of the linear feature.
Fig. 3
Fig. 3. Galaxies on the DF2–DF4 axis.
Legacy survey image of the central region of the NGC 1052 group, highlighting the 11 galaxies that are part of the trail according to the Hough transform. Several (2 ± 2) of these are expected to be chance projections. LEDA 4014647 is a candidate for an interloper (that is, an unrelated group member) given its brightness and relative compactness. Its radial velocity was listed as 1,680 ± 60 in earlier SDSS releases (Data Release 3) but was later erroneously revised to a z = 0.7 quasi-stellar object (Data Release 16). Judging from morphology alone, RCP 26 and Ta21-12000 may also be chance projections. Besides DF2 and DF4, RCP 32, DF5 and DF7 all satisfy or nearly satisfy the ultra-diffuse galaxy criteria. RCP 32 and DF7 are candidates for the two dark-matter-dominated remnants that have been predicted to precede dark-matter-deficient galaxies along the post-collision trajectory. We also highlight DF9 (SDSS J024007.01−081344.4), a galaxy with a bright star cluster that falls on the trail but is not part of the objectively selected sample.
Fig. 4
Fig. 4. The sizes of galaxies on the DF2–DF4 axis.
a, The apparent size–apparent magnitude relation for low-surface-brightness objects in NGC 1052, using recent measurements from Legacy data. Galaxies that are part of the trail are labelled with solid symbols. The thin solid line is a running median, with N = 7. The dashed line is a least-squares fit to the running median and has the form log re = −0.09(g − 20) + 0.87 kpc. Galaxies in the trail are typically larger than other galaxies of the same magnitude. b, Distribution of galaxies colour-coded by their location with respect to the least-squares fit. The largest galaxies in the group are preferentially located along the DF2–DF4 axis.
Extended Data Fig. 1
Extended Data Fig. 1. Illustration of the proposed formation scenario of DF2 and DF4.
Two gas-rich dwarf galaxies experience a high-speed encounter with a small impact parameter (top). Following previous studies the collisional gas gets stripped and shocked at closest approach, and forms stars at a prodigious rate with a bias towards massive clumps (second from top). The dark matter and previously formed stars are tidally distorted but continue ahead of the newly forming galaxies (third from top). Feedback in the absence of a dark-matter halo leads to expansion of the newly formed galaxies. Most of the newly formed stellar mass is in two clumps, but several lower mass galaxies have also formed in the wake (bottom).
Extended Data Fig. 2
Extended Data Fig. 2. Morphology of the NGC 1052 group.
Distribution of 72 probable group members from a recent compilation of galaxies in the NGC 1052 field. Contours were derived with the non-parametric kernel density method. The dashed line indicates the trail. There is no evidence that the trail is in the general direction of large-scale structure in this field.
Extended Data Fig. 3
Extended Data Fig. 3. A candidate dark-matter-dominated galaxy.
HST/ACS images of DF7, which is located at the western end of the galaxy trail, beyond DF4. Left: a colour image generated from the F606W and F814W data. Right: a median-filtered version of the F606W image, with the arrow depicting the direction towards DF4. DF7 is a candidate for the dark-matter-dominated remnant of one of the original galaxies, given its location and its elongation in the direction of DF4.

Comment in

References

    1. Cohen Y, et al. The Dragonfly Nearby Galaxies Survey. V. HST/ACS observations of 23 low surface brightness objects in the fields of NGC 1052, NGC 1084, M96, and NGC 4258. Astrophys. J. 2018;868:96.
    1. van Dokkum P, et al. An enigmatic population of luminous globular clusters in a galaxy lacking dark matter. Astrophys. J. 2018;856:L30.
    1. van Dokkum P, Danieli S, Abraham R, Romanowsky AJ. A second galaxy missing dark matter in the NGC 1052 group. Astrophys. J. 2019;874:L5.
    1. Dutta Chowdhury D, van den Bosch FC, van Dokkum P. On the orbital decay of globular clusters in NGC 1052-DF2: testing a baryon-only mass model. Astrophys. J. 2019;877:133.
    1. Dutta Chowdhury D, van den Bosch FC, van Dokkum P. On the evolution of the globular cluster system in NGC 1052-DF2: dynamical friction, globular-globular interactions, and galactic tides. Astrophys. J. 2020;903:149.

Publication types