The new dark matter map reveals hidden bridges between galaxies.
Astronomers from the University of Pennsylvania with colleagues from other countries noted that a map of dark matter was previously created: it was made using the latest machine learning technologies. Some of the structures visible on the map were already known to science, but some were discovered for the first time.
Dark matter in astronomy and cosmology, as well as in theoretical physics, is a form of matter that does not participate in electromagnetic interaction and therefore is inaccessible to direct observation. It is about a quarter of the mass-energy of the Universe and manifests itself only in gravitational interaction.
In their new work, the researchers used a set of models of galaxies Illustris-TNG, which contains data, including on dark matter, on the basis of which AI was trained. The result is a model that can predict the distribution of dark matter clumps in the universe. By applying the model to data from the Cosmicflow-3 galactic catalog, the researchers obtained a map of the distribution of dark matter.
The map from our models does not match the simulation data perfectly, but we can still reconstruct very detailed structures. We found that including the motion of galaxies – their radial velocities – in addition to their distribution, significantly improved the quality of the map and allowed us to see these details.
As a result, the authors found that such matter connects neighboring galaxies with invisible thread-like structures – a kind of dark matter bridges. Studying how the distribution of dark matter relates to other data will help us understand the nature of dark matter.