Galaxies with no dark matter?

Astrophysicists from Yale University have recently claimed the discovery of a galaxy lacking dark matter ¹. But, to fully understand (and assess) the excitement produced by this paper, let’s review a couple of things regarding dark matter and galaxies. In the current lambda-Cold Dark Matter (ΛCDM) paradigm, the most accepted “theory to model the Universe in large scales”, baryonic (visible) matter assembles where dark matter is denser. For this reason, it is expected that all visible galaxies reside embedded within a dark matter halo. However, it has been known for some time now that this is not the case for all galaxies. The origin of some of them is related to merger events, being the outcome of the collapse of gas and dust that have been flung out of merging systems. As a consequence, the galaxies formed in this manner should end up having no dark matter at all. These galaxies are named “tidal dwarfs” ², and not only have they been found observationally, but are also the solution to explain, for instance, why some galaxies such as the Milky Way, M31 or CentaurusA have satellites (dwarf galaxies) orbiting in a plane around them (³, aspect that has been previously used against ΛCDM). Having said this, a galaxy lacking dark matter does not seem so strange, does it?

NGC1052-DF2 is a faint, extended galactic system located in projection in the NGC1052 group of galaxies. Such faint object is only observable using space photometry (Hubble Space Telescope, see figure 1) or wide field observations optimized to detect low surface brightness structures. Other observations only reveal an over-density of point-like sources where the galaxy is located, over-density that has been associated with the globular cluster system of NGC1052-DF2. An analysis of the photometric properties of the galaxy, together with a careful study of how those globular clusters move around it, have allowed van Dokkum’s team to conclude that this galaxy has a very low amount of dark matter around it, being compatible with an absolute absence.

These findings are compatible with NGC1052-DF2 being i) a tidal dwarf, ii) consequence of a complex interaction between matter and quasar winds, or iii) the remnants of disruption of material falling towards NGC1052. In addition, contrary to what might be expected, the discovery of galaxies with no dark matter strongly supports the well-established ΛCDM paradigm: Alternative theories as MOND (Modified Newton Dynamics) will fail at reproducing the properties of a galaxy with no dark matter. The explanation is simple. In ΛCDM you can avoid the effect of dark matter by simply not adding it, whereas in MOND, where the physics has been modified to consider the effect of such elusive matter, you will always have its effect. Until this point everything seems fair and interesting. However, strong claims require even stronger evidences, and here is where this work is causing great controversy in the astrophysical community.

The idea behind this study is “simple”. On the one hand, by knowing the light that is emitted by the system, we can obtain the baryonic mass, the amount of normal/visible matter that composes the galaxy and that is emitting such light. This way, the authors state that NGC1052-DF2 has around 2×10⁸ solar masses. On the other hand, by inspecting how the system of globular clusters orbits around it (see figure 2), we can infer the total amount of mass, i.e. the mass that affects the movements of such globular clusters, including baryonic and dark matter. By doing this, the authors find that this galaxy should have a dark matter halo less massive than 1.5×10⁸ solar masses, compatible with having few (or even an absence) of dark matter.

However, there are subtle but important aspects of the author’s analysis that deserve some clarification and that are making astrophysicists all around the world to revisit their results. First, the derivation of the amount of baryonic matter in this system relies on its distance. In Astronomy, everything is projected into the celestial sphere. Depending on the distance to the objects, they might be extremely small and located in our vicinity or extremely large and located far away from us, having in both options the same angular size. A plane and a bird do look similar in the sky, don’t they? The authors assume a distance to NGC1052-DF2 of 20 Mpc, whereas other distance estimators (also shown by the authors) suggest that the system is only 10 Mpc away from us. Such factor of two would move NGC1052-DF2 from a galaxy lacking dark matter to a normal dwarf galaxy. In addition, assuming a distance of 20 Mpc would imply that all the globular clusters surrounding it would be similar to ω Centauri, the largest globular cluster in our Milky Way, something that is quite unlikely. Secondly, a different source of caution comes from the derivation of the dark matter (dynamical matter). For doing that, the authors are using the kinematics of just 10 systems located all around the target. The data analysis of such small sample is full of uncertainties, uncertainties that once more, would move our target from a galaxy lacking dark matter to a normal dwarf system (although we must admit that in this sense the authors did quite a thorough job).

This articles is not intended as criticism towards this interesting discovery, but rather as a reflexion on how science works. Science is a collaborative effort in which every member contributes their bit. Trial and error and working as a community (paying attention to what other scientists think about our work) are what have made it possible to know what we know about the Universe around us. Astronomy is no exception.