What is dark matter, and how do we know it exists?

Dark matter is a term for matter that can be inferred from its gravitational effects but does not emit or absorb detectable amounts of light. NASA says scientists estimate that ordinary matter makes up about 5% of the universe, dark matter about 27%, and dark energy about 68%. Dark matter does interact with ordinary matter through gravity, but it does not appear to interact with the electromagnetic spectrum, so it does not absorb, reflect or emit light. Its existence is widely accepted, although its identity is not. [1] [2]

In 1933 the astronomer Fritz Zwicky found that galaxies in the Coma cluster moved too quickly for the gravity of their visible matter to hold them. They should have been escaping the cluster, but they stayed together, which implied extra unseen mass. [1] [3]

In the 1970s Vera Rubin studied stars at the outer edges of spiral galaxies. They moved so fast that a large amount of unseen matter seemed necessary to keep them from flying off into intergalactic space. NASA describes this work as strong enough evidence that the scientific community embraced the concept. Theoretical arguments also showed that disk galaxies and the Local Group need significant dark matter halos to remain dynamically stable. [1] [3]

Gravity bends light, so mass that cannot be seen still distorts the images of distant galaxies behind it. This gravitational lensing lets astronomers map where the mass is. A 2015 review in a scientific journal says lensing by galaxies, clusters and large-scale structure directly confirmed dark matter and measured its distribution. [1] [3]

The Bullet Cluster, formed when two galaxy clusters collided, is often cited as some of the best direct evidence. Hot gas made of normal matter, seen in X-rays by Chandra, lies in a different place from most of the mass, which lensing reveals. Researchers think that mass is likely dark matter. [1]

Measurements of the cosmic microwave background and of light-element abundances point the same way. The review gives a total matter density of about 25% of the critical density, against only about 4.5% for ordinary (baryonic) matter. That gap means dark matter must be nonbaryonic, so it is not made of ordinary atoms. Dark but ordinary objects such as planets and stellar remnants are counted in the baryon figure, not as dark matter. [3]

The data indicate that dark matter is cold, meaning it moved slowly in the early universe, and that it interacts only weakly with matter other than through gravity. Such a particle would require physics beyond the Standard Model. Many candidates have been proposed, including hypothesized partner particles to known ones, but the 2015 review says none has been detected. Experiments for direct and indirect detection continue. [3] [4]

A minority approach holds that gravity itself behaves differently on large scales, which would remove the need for dark matter. Modified Newtonian dynamics is one example. The review says such models are not completely ruled out but have serious difficulty fitting precise data, especially from galaxy clusters like the Bullet Cluster and from the cosmic microwave background. The dark matter picture is the one the review treats as fitting the accumulated data remarkably well. [3]