Two dark matter-less galaxies shed light on ghostly galaxies that are usually dark matter rich.
Ghosts are everywhere — ghost galaxies, that is. Some are the size of the Milky Way or even bigger, yet they have so few stars that they’re extremely faint, virtually see-through, and difficult to detect. We’ve known such galaxies exist for decades, but only recently have new observing techniques shown just how common they are.
Most of these so-called ultra-diffuse galaxies must be unusually rich with dark matter — it would provide the gravitational pull to keep the sparse stars from disbanding. But now astronomers have found not just one, but two such galaxies that appear to lack dark matter altogether. The new discovery complicates the picture of how these galaxies came to be.
A Dark Matter-less Galaxy
Last year, Pieter van Dokkum (Yale University) and colleagues discovered the first dark matter-less galaxy on the outskirts of the elliptical NGC 1052, dubbed NGC 1052-DF2. Observing 10 globular clusters looping around the galaxy’s edges, the astronomers determined DF2’s total mass. Turns out, the total mass is tiny and roughly equivalent to its mass in stars. Usually, a galaxy’s total mass outweighs its stars by a factor of 100 or more, hinting at the presence of dark matter. But in this galaxy, there doesn’t seem to be any hidden matter at all — just stars.
(For reasons astronomers still don’t understand, the globular clusters themselves are odd, too. They’re all unusually luminous, about as bright as Omega Centauri, the brightest globular cluster in the Milky Way.)
It’s perhaps no surprise that the team’s extraordinary claim was immediately surrounded by controversy. Some astronomers argued that 10 globular clusters were not enough to give a reliable mass estimate. Others argued that the distance to NGC 1052-DF2 was itself uncertain, which put any attempts to understand it in disarray.
To characterize DF2, van Dokkum and colleagues had used the Dragonfly Telephoto Array, which images each target simultaneously through multiple Canon 400mm lenses, akin to how a dragonfly’s eye works. The unique setup enables the astronomers to catch extremely faint “fuzzies,” such as large, star-poor galaxies. Before the discovery of DF2, the group had characterized dozens of these ghostly galaxies in the crowded Coma Cluster, including the curious case of Dragonfly 44, a Milky Way-mass galaxy with 100 times fewer stars than our galaxy. It appears to be made of 99.99% dark matter.
Van Dokkum and colleagues had suggested that, as galaxies like Dragonfly 44 fly through the teeming cluster environment, gravitational interactions strip away the stars or star-forming material. These ultra-diffuse objects, then, are essentially “failed galaxies.”
The discovery of DF2, though, seems to turn that argument on its head — something has stripped this galaxy of its dark matter, or maybe it never had any to begin with.
“One Is an Exception, Two Is a Population”
Now, van Dokkum’s team has discovered a new dark matter-less galaxy: NGC 1052-DF4. After finding DF4 in Dragonfly images, the team followed up with Hubble Space Telescope imaging and spectroscopy through the Keck I telescope on Mauna Kea, Hawai‘i.
DF4 is DF2’s doppelgänger: It’s roughly the same size, luminosity, and it, too, has a collection of unusually bright globular clusters. Also like DF2, DF4 appears to have a total mass equivalent to its stellar mass, once again negating the need for dark matter.
“What the paper can't convey is how incredibly surprised we were!” van Dokkum wrote on Twitter shortly after the paper’s release on the astronomy preprint arXiv. The study appears in the March 20th Astrophysical Journal Letters.
The discovery of a second dark matter-less galaxy lends credence to the first find, and not only because two are greater than one. In a case of peer review going spectacularly right, van Dokkum and colleagues followed an anonymous referee’s suggestion to measure DF4’s mass a second way, by measuring the velocity of the stars in the galaxy itself. Neither Dragonfly, Hubble, nor Keck can resolve the faint, faraway stars into points; instead, the astronomers used Keck to pick up the stars’ diffuse light, much like stargazers under dark skies can see the faint, milky glow of billions of stars in our galaxy.
By measuring the velocities of the diffuse starlight, the astronomers confirmed the mass they had estimated before. The team, this time led by Shany Danieli (Yale), also went back and measured the diffuse light from DF2, confirming its low mass as well. The updated results on DF2 are published in the April 1st Astrophysical Journal Letters (preprint available here.)
“I think there is truly something to the low velocity dispersion (and therefore low mass) claims,” says Michelle Collins (University of Surrey, UK), who was not involved in the study. She had voiced doubts about DF2 when it was first discovered, but she notes, “I’m certainly not worried about the statistics any more.”
Nevertheless, there’s still work to be done. A lot of circumstantial evidence suggests that DF2 and DF4 belong to the cluster of galaxies around NGC 1052, but not everybody is convinced.
“The outstanding issue,” Collins continues, “is the distance to both DF2 and DF4. If they were truly closer, the galaxies would appear more typical, and would have a typical amount of dark matter. Without more data, this distance estimate remains highly uncertain, so I’m honestly not sure what to make of it.” Hubble observations coming up in July or August should decide the distance question once and for all.
Dark Matter-Rich Galaxies
So why do NGC 1052-DF2 and DF4 look so different from most of the ultra-diffuse galaxies seen nearby in the Coma Cluster? Among the thousands of like galaxies lurking there, most are rich in dark matter, with up to 98% of their mass being “dark.”
“We may just be highlighting the extreme ends of the distribution,” Collins says. “In between, I suspect there will be a number of ultra-diffuse galaxies that have roughly typical amounts of dark matter, too.”
DF2 and DF4 do have one common characteristic with the dark matter-rich Coma Cluster galaxies: a preponderance of globular clusters. In 2017 van Dokkum and colleagues found that more than half of the ultra-diffuse galaxies in Coma had more globular clusters than expected. These ancient stellar cities were born billions of years ago, and they probably hold clues as to the galaxies’ birth and evolution — still, it’s not clear how the galaxies and their globular clusters are related.
To try to make sense of ultra-diffuse galaxies as a whole, Pavel Mancera Piña (ASTRON and University of Groningen, The Netherlands) and colleagues recently undertook a large survey of eight galaxy clusters, finding 442 galaxies that meet the qualifications of being both large and very faint. The group found that more of these galaxies tend to be “red and dead” when they’re closer to the cluster centers, which suggests that the cluster environment has torn away the material needed to make new (and blue) stars.
The study, published in the May issue of the Monthly Notices of the Royal Astronomical Society, suggests that ultra-diffuse galaxies are essentially big dwarfs: They have normal amounts of stars for dwarf galaxies, it’s just that they’re unusually large. That explanation runs counterpoint to the “failed galaxy” theory that van Dokkum and others have espoused.
“The definition of ultra-diffuse galaxies is very broad,” Collins notes. “Given the huge range of properties allowed by the current definition, I think it’s likely there is more than one ‘type’ of galaxy caught in this definition.”
But galactic ghosthunters haven’t given up: As Piña’s group continues a series of studies on the ultra-diffuse population as a whole, the Dragonfly team is still searching for the dark matter-less outliers.