Astronomers have tested a method for reconstructing the state of the early Universe by applying it to 4000 simulated universes using the ATERUI II supercomputer at the National Astronomical Observatory of Japan (NAOJ).
They found that together with new observations the method can set better constraints on inflation, one of the most enigmatic events in the history of the Universe. The method can shorten the observation time required to distinguish between various inflation theories.
Just after the Universe came into existence 13.8 billion years ago, it suddenly increased more than a trillion, trillion times in size, in less than a trillionth of a trillionth of a microsecond; but no one knows how or why.
This sudden “inflation,” is one of the most important mysteries in modern astronomy. Inflation should have created primordial density fluctuations which would have affected the distribution of where galaxies developed. Thus, mapping the distribution of galaxies can rule out models for inflation which don’t match the observed data.
However, processes other than inflation also impact galaxy distribution, making it difficult to derive information about inflation directly from observations of the large-scale structure of the Universe, the cosmic web comprised of countless galaxies. In particular, the gravitationally driven growth of groups of galaxies can obscure the primordial density fluctuations.
A research team led by Masato Shirasaki, an assistant professor at NAOJ and the Institute of Statistical Mathematics, thought to apply a “reconstruction method” to turn back the clock and remove the gravitational effects from the large-scale structure.
They used ATERUI II, the world’s fastest supercomputer dedicated to astronomy simulations, to create 4000 simulated universes and evolve them through gravitationally driven growth.
They then applied this method to see how well it reconstructed the starting state of the simulations. The team found that their method can correct for the gravitational effects and improve the constraints on primordial density fluctuations.
“We found that this method is very effective,” says Shirasaki. “Using this method, we can verify of the inflation theories with roughly one tenth the amount of data. This method can shorten the required observing time in upcoming galaxy survey missions such as SuMIRe by NAOJ’s Subaru Telescope.”
- 38Japanese astronomers have developed a new artificial intelligence (AI) technique to remove noise in astronomical data due to random variations in galaxy shapes. After extensive training and testing on large mock data created by supercomputer simulations, they then applied this new tool to actual data from Japan's Subaru Telescope and…