An artificial intelligence model can now detect and predict the hidden turbulent motions taking place inside the Sun’s atmosphere. Developed by a team of researchers in Japan, this neural network has been fed with vertical motion and temperature data that was collected from the surface of the solar photosphere. The solar photosphere is the surface of the Sun and is also considered the lowest layer of the solar atmosphere. The solar photosphere is the surface where solar flares, coronal mass ejections, and sunspots originate and occur. On looking closely, one can figure out that the solar photosphere is not uniform. Granules, the top layers of convection cells in the solar plasma, make up the variable surface of the Sun.
The researchers have developed a novel system of convolutional neural networks that estimates the spatial distribution of horizontal velocity by factoring in the spatial distribution of temperature and vertical velocity. “Our network exhibited a higher performance on almost all the spatial scales when compared to those reported in previous studies,” said Ryohtaroh Ishikawa, lead author of the study, as reported by Science Alert.
Along with the team, Ishikawa developed a numerical simulation to train their neural network. During the training, the researchers used three different sets of simulation data. The team found that the AI developed by them can function rather accurately with just temperature and vertical flow data. It can accurately predict horizontal motion.
While the neural network developed by Ishikawa and the team was accurate in predicting and detecting large-scale flows, picking small-scale turbulence was not that easy for the AI. “This implies that the network was not appropriately trained to reproduce the velocity in small scales generated by turbulent cascades,” the paper read.
With the neural network in action and developing, researchers believe that the data collected from the AI can deepen the understanding of solar convection and processes that generate jets and explosions erupting from our fascinating and forbidding star.