A neural network can predict hidden turbulent motion inside the Sun.
The data collected from the surface of the solar photoosphere was fed into the model. This could help us understand the processes that cause explosions and jets to erupt from the Sun.
A team of researchers led by astronomer Ryohtaroh Ishikawa of the National Astronomical Observatory of Japan developed a novel neural network to estimate the spatial distribution of horizontal velocity.
This led to an efficient detection of spread features. When compared to previous studies, our network exhibited a higher performance on almost all the spatial scales.
The solar photoosphere is the region of the Sun's atmosphere that is commonly referred to as its surface. The lowest layer of the solar atmosphere and the region in which solar activity such as sunspots, solar flares and coronal mass ejections originate.
The surface of the photoosphere is not uniform. It is covered with sections crowded together, lighter in the middle and darker towards the edges. The tops of the cells in the solar plasma are called granules. The hot plasma rises in the middle and then falls back down around the edges as it cools.
When we observe these cells, we can measure their temperature, but not their horizontal motion. Smaller scale flows in these cells can cause other solar phenomena. Scientists want to understand exactly how the photoosphere behaves, because turbulence is thought to play a role in heating the solar corona.
Ishikawa and his team used simulation data to train their neural network. They found that the artificial intelligence could accurately describe horizontal flows in the simulations that would be invisible on the real Sun.
We could feed it solar data and expect the results to be in line with what is happening on our star.
The neural network needs some tweaking. While it was able to detect large-scale flows, it had trouble picking out smaller features. The researchers said that resolving small-scale turbulence should be the next step in the development of their software.
By comparing the results of the three models, we found that the decrease in coherence spectrum occurred on the scales that were lower than the energy injection scales. They wrote in their paper that the network was not trained to reproduce the velocity fields in small scales generated by turbulent cascades.
These challenges can be explored in future studies.
The researchers are working on a software that will help better understand turbulence in fusion plasmas.
The research has been published.
All Rights Reserved © 2024
