Since the first theory in the 1970s, astrophysicists and cosmologists have done their best to solve the mystery of dark matter. It is believed that this invisible mass makes up 85% of the matter in the universe and makes up 27% of its mass-energy density. In addition, it also provides the large-scale skeletal structure of the universe (the cosmic web), which, due to its gravitational influence, determines the movements of galaxies and material.
Unfortunately, the mysterious nature of dark matter means that astronomers cannot study it directly, which prevented it from measuring its distribution. However, it is possible to infer its distribution based on the observable influence of its gravity on local galaxies and other celestial objects. Using state-of-the-art machine learning techniques, a team of Korean-American astrophysicists was able to create the most detailed map of the local universe to date, showing what the “cosmic web” looks like.
The team responsible for this breakthrough was led by senior researcher Sungwook E. Hong of Seoul University and the Korea Astronomy and Space Science Institute (KASI). He was joined by the associate professor Donghui Jeong from the Institute for Gravitation and the Cosmos (IGS) at Penn State and the researchers Ho Seong Hwang and Juhan Kim from the Seoul National University and the Korea Institute for Advanced Study (KIAS).
In the past, previous attempts to map the cosmic web began with a model of the early universe and then simulated its evolution over billions of years. However, this method has had limited success because of the enormous amount of computing power required. Taking a different approach, the team created a model that used machine learning to predict the distribution of dark matter based on the known distribution and movement of galaxies.
The team built and trained this model using Illustris-TNG, a cosmological project that has performed multiple simulations using galaxies, gases, other forms of baryonic (also known as visible) matter, as well as dark matter. The team selected simulated galaxies from Illustris-TNG that were comparable to the Milky Way and identified the properties needed to predict the distribution of dark matter. Said Jeong:
“Ironically, it is easier to study the dark matter distribution much further away because it reflects the very distant past, which is much less complex. Over time, as the large-scale structure of the universe has grown, the complexity of the universe has increased, making it inherently more difficult to take measurements locally on dark matter. “
“Once the model receives certain information, it can essentially fill in the gaps based on what it has considered previously. The map from our models does not perfectly match the simulation data, but we can still reconstruct very detailed structures. We found that the inclusion of the motion of galaxies – their own radial velocities – in addition to their distribution, dramatically improved the quality of the map and allowed us to see these details. “
Map of the distribution of dark matter within the local universe using a model to infer its location due to its gravitational influence on galaxies. Photo credit: Hong et. al., Astrophysical Journal
The next step was to apply this model to real data from the local universe that the team had obtained from the Cosmicflow-3 database. This astronomical catalog contains extensive data on the distribution and movement of over 17,000 galaxies in a 650 million light year region (200 megaparsecs) around the Milky Way. The resulting map successfully reproduced well-known prominent structures in the local universe.
These included the “Local sheet, “Region of space that contains the Milky Way, Andromeda (and other members of the“ local group ”), and the galaxies of the Virgo Cluster. Another outstanding structure was the “Local emptiness, “A relatively empty space region next to the local group. In addition, several new structures have been identified on the map, such as: B. smaller filament structures that act as hidden connections between galaxies.
As you can see from the cross-sections of the map (see above), large concentrations of luminescent material are shown in red, while largely empty sections are shown in blue. Galaxies are referred to as small black dots, the Milky Way is denoted by the black X in the center, and the arrows represent the movement of these large-scale structures. These connecting filaments, which appear as wispy yellow threads, need to be reexamined to learn more about these previously unknown features. Said Jeong:
“A local map of the cosmic web opens a new chapter in cosmological investigation. We can study how the distribution of dark matter affects other emission data, which helps us understand the nature of dark matter. And we can examine these filament structures directly, these hidden bridges between galaxies. “
“Because dark matter dominates the dynamics of the universe, it basically determines our fate. So we can ask a computer to evolve the map for billions of years to see what will happen in the local universe. And we can develop the model further in time to understand the history of our cosmic neighborhood. “
Illustris simulation showing the distribution of dark matter in 350 million by 300,000 light years. Galaxies are represented as high-density white dots (left) and normal baryonic matter (right). Photo credit: Markus Haider / Illustris
For example, scientists have known for some time that the Milky Way and Andromeda galaxies are slowly approaching. However, whether or not they will collide to form a supergalaxy (uncreative nickname Milkomeda) in an estimated 4.5 billion years remains unclear. By studying the filaments of dark matter that connect our two galaxies, astrophysicists could gain valuable insights into their future.
Hong and his colleagues also plan to improve the accuracy of their map by adding more galaxies. This will be possible thanks to next-generation missions like the James Webb Space Telescope (JWST), which will finally launch into space on October 31, 2021. With its advanced suite of instruments, the JWST will examine the universe in the long-wave range, visible and near-infrared to mid-infrared wavelengths.
In this way, astronomers can identify galaxies that are smaller, weaker, and further away from our solar system. Improvements in computing and machine learning will also result in bigger and better simulations that can make out more galaxies over longer periods of time. Similarly, missions such as ESA’s Gaia Observatory provide more accurate data on the correct movements and speeds of galaxies (astrometry).
The planned successor, ESA’s Euclid Observatory, is scheduled to start in 2022 and collect data on two billion galaxies in 10 billion light years of space. This will be used to create the most detailed 3D map of the local area of the universe to date, which is intended to provide important clues as to the role of dark matter (and dark energy) in cosmic evolution. These maps provide a means of comparison for astronomers to know that their physics models are spot on.
The study describing its results, “Uncovering the deep cosmic web of galaxies through deep learning,” recently appeared in the Astrophysical Journal. This research was made possible with support from the National Research Foundation of Korea, funded by the Korean Department of Education, Korea Department of Science, the US National Science Foundation (NSF), NASA Astrophysics Theory Program, and the KIAS ‘Center for Advanced Computation.
Further reading: PSU, The Astrophysical Journal