A view of the cosmic web as re-created in a cosmological simulation. Simulations like these were compared to telescope observations to determine the "fuzziness" of the dark matter. Source: Chris Pedersen


There are a few candidates for dark matter, the mysterious substance that today accounts for over four-fifths of the Universe's mass. In the fuzzy dark matter model dark matter is composed of very low-mass particles called ultralight axions which are a million trillion trillion times lighter than the simplest atoms. The idea comes from string theory, and may explain possible gaps in our understanding of how galaxies form. However, this new study rules out ultralight axions based on the way that their presence would make our Universe look.

The model predicts that fuzzy dark matter would completely erase structure in the Universe below a certain length-scale, set by its quantum wavelength, or fuzziness. This study looks for this clear signature in an astrophysical phenomenon called the Lyman-alpha forest which is composed of many absorption features created by clouds of gas seen along the line-of-sight to distant galaxies. The research looked at the Lyman-alpha forest probing smaller length-scales than ever before by improving their simulations, and did not see the tell-tale signal of fuzzy dark matter. This rules out the best-motivated formulation of fuzzy dark matter with high statistical significance, meaning that there is a vanishingly-small chance that all the dark matter is made up of these ultralight particles.

"Our study rules out the most popular model of fuzzy dark matter but there is much work still to do. We now have the tools to test other ideas as to what is the dark matter. Meanwhile, heavier axions (called QCD axions) remain viable and work is being pioneered at the OKC to try to detect these also. There is a huge community effort to develop new techniques like we applied here to aid our search" says former OKC Fellow Keir Rogers.

Sophisticated machine learning methods were used to carry out the study. These methods can now be used to test other ideas for the composition of dark matter. This research was conducted as part of the AxionDM research environment at Stockholm University.