Concentrations of dark matter (mapped in contours above) usually - but not always - match up with normal matter (coloured).Hot on the heels of evidence last year that dark matter really does exist (see 'Dark matter spied in galactic collision'), the same technique has been used to map this uncharacterized mass across half a million distant galaxies.
The map shows that, as predicted, the mysterious dark matter that makes up a quarter of the Universe forms a filamentous 'skeleton' upon which visible matter congregates, eventually producing stars. This is the first time such a large-scale three-dimensional picture of dark matter has been produced, and it will allow cosmologists to probe deeper into the nature of this elusive matter.
But the map, published in Nature1, also has a few puzzles within it. Some areas show clumps of dark matter that aren't accompanied by the bright features associated with conventional, visible material (made of baryonic matter), and vice versa.
"On the large scale the general picture is as expected, but there are some small-scale discrepancies," says Richard Massey at the California Institute of Technology, Pasadena, and one of the team members who pieced together the map from hundreds of slightly overlapping images from the Hubble Space Telescope's Cosmic Evolution Survey (COSMOS).
The existence of large clumps of isolated dark matter and visible matter flies in the face of everything we know, says cosmologist Carlos Frenk of the University of Durham, UK.
The discrepancies could be a simple error resulting from the way the observations were made. But if they are real, says Massey, they will bring a huge shock. "Baryonic structures are expected to form only inside the dark-matter scaffold," he says. "There will need to be a lot of follow-up work before we really believe any individual discrepancies."
A light pull
Massey used a technique called gravitational lensing, whereby the pull from dark matter caught in between a star and the observing telescope alters the path of the light, and allows the presence of dark matter to be inferred.
Eric Linder of the University of California, Berkeley, who was not involved with the work, agrees that the map backs up the favoured theory that dark matter forms a scaffold on which galaxies form.
He suggests possibilities for the more unusual spots in the map: one is that galaxies made of dark matter (dark galaxies) exist, but he thinks this is unlikely. Another possibility is that the discrepancies are errors in the data — which seem almost inevitable given that mapping the dark matter required a very sensitive measurement of an incredibly small signal. "Right now the discrepancies are curiosities rather than items of concern," Linder says.
Massey is also confident in the robustness of his map on the whole. "A couple of individual discrepancies in the map are not a huge surprise," he says. "The technique is intrinsically more noisy, and more prone to systematic errors, near the edges of the map." That is where most of the discrepancies are seen.
Blown away
ADVERTISEMENT
There are plausible explanations for small areas of dark matter and visible matter existing in isolation.
Dark matter, if the clump is small enough, could have any accumulating visible matter blown out of it by a high-energy phenomenon such as a quasar or a supernova, for example. The collision of two galaxies could also blow an amount of visible matter out as a faint satellite galaxy that has no associated dark matter, suggests Frenk.
But these theories can't explain the large features visible on the COSMOS map, he adds.
Like Massey and Linder, Frenk also suspects that the discrepancies are due to errors: "We know too much about the Universe," he says, to have completely missed this phenomenon up till now.
Visit our mattermapped.html">newsblog to read and post comments about this story.
© 2019 Nature is part of Springer Nature. All Rights Reserved.
