How do you put universe hypotheses to the test? By constructing massive supercomputers and simulating the evolution of the universe.
A group of Japanese researchers has created the world’s largest cosmic simulation, which includes microscopic “ghost” particles known as neutrinos. The researchers employed a remarkable 7 million CPU cores to solve for the evolution of 330 billion particles and a computational grid of 400 trillion units to investigate one of physics’ largest unresolved puzzles.
Dark matter is by far the most important type of matter in the cosmos. We have no idea what it is or what it is composed of, but there is a lot of it. It accounts for around 80% of all matter. Baryonic matter, which makes up stars, planets, and the entire periodic table’s rich variety, makes up a tiny percentage of all the matter in the universe.
The universe’ backbone is made up of dark matter. There were no structures in the universe billions of years ago. All of the substance, black or light, was evenly dispersed and not lumpy in the least. There were few differences in density from one location to the next. In general, it was a rather dull universe.
However, as time passed, the universe grew more fascinating. There were minuscule density discrepancies in the early seconds of the Big Bang, which were seeded by microscopic quantum fluctuations. Dark matter began to gather together in places with slightly higher density and slightly more gravity. As those early buildings grew, additional material was attracted to them. This process emptied enormous parts of the cosmos — now known as cosmic voids — over billions of years, pushing all matter into a vast network of clusters, walls, and filaments.
Then there are neutrinos, which are extremely small particles with almost no mass. Indeed, they account for less than 0.1 percent of the universe’s total mass. These microscopic particles, on the other hand, have a huge impact on the evolution of structures. They’re quick – incredibly fast — and can travel at speeds approaching those of light. The creation of massive structures like galaxies and clusters is slowed by this extraordinary speed.
Neutrinos are too quick to settle down in one place, whereas dark matter wants to keep stacking up through gravity. Despite the fact that neutrinos have extremely little mass, they do have some. They can use gravity to gently control dark matter’s behavior, preventing it from clustering as tightly as it might otherwise.
To put it another way, the cosmos is a little smoother than it would be if neutrinos didn’t exist.
Universe’s Unsolved Mysteries
A key unsolved problem in current physics is determining the masses of the three known neutrino “flavors” – electron neutrinos, muon neutrinos, and tau neutrinos. Ironically, we can determine the masses of these tiny particles by studying the universe’s greatest structures.
Cosmologists frequently use computer simulations to try to understand the nature of dark matter and the role of neutrinos in determining cosmic evolution. If the neutrino mass is changed slightly in the simulations, the neutrinos’ influence on the creation of structures over billions of years will alter. So you can figure out how much neutrino mass there is by measuring those identical structures.
These simulations typically cover only a small portion of the real universe and begin with a collection of dark matter “particles,” each of which represents a specific amount of dark matter — for example, a single blob with a mass millions of times that of the sun. These particles are then placed in the simulations as they would be in the early cosmos. The models follow how those particles evolve as a result of their reciprocal gravitational attraction, resulting in the massive structures we see today.
Because the true behavior of dark matter can only be represented by a small number of particles, this is an approximation technique, but it works well for dark matter. Because of their incredible speed, simulating neutrinos is significantly more difficult. Because they can go from one side of the simulation to the other in a short length of time, it’s tough to follow their behavior within the simulation. As a result, the simulations can’t keep up with how neutrinos behave and influence dark matter.
It’s a math problem
So perhaps we shouldn’t bother attempting to approximate neutrino behavior. To correctly follow the evolution of neutrinos and account for their rapid behavior, an extremely complex equation must be solved. However, solving this equation, known as the Vlasov equation after Russian scientist Anatoly Vlasov, necessitates massive computing power.
So a group of Japanese scientists employed the Fugaku supercomputer’s 7 million processors to track the evolution of dark matter and the influence of neutrinos on structure formation. In the largest simulation of its type, the researcher employed 330 billion particles to represent dark matter and a 400 trillion component computing grid to depict neutrinos.
While the simulation did not explain the puzzle of neutrinos’ mass, it did pave the door for more of its sort. This simulation was essentially a proof-of-concept to demonstrate that we can now include neutrinos in simulations with more accuracy than ever before. Future simulations, armed with this new technology, will provide insight into the role of neutrinos in the universe, and possibly even reveal a key to unlocking their mass.