Large solar storms can cause widespread power outages and permanent damage to communication systems on Earth. Currently, our understanding of their formation is limited, and predictions are challenging. Researchers from the University of Hawaii have developed a new method to improve predictions using modern solar observation techniques combined with computer science and data from the world’s largest solar telescope located on the 3,000-meter-high Haleakala volcano in Maui.
The goal is to quickly analyze massive amounts of data to better understand the solar atmosphere and the formation of solar storms. This involves using simulations and modern AI methods for analysis. Every day, dozens of terabytes of data are collected that need to be analyzed. According to astronomer Kai Yang from the University of Hawaii, combining this data with machine learning offers a valuable opportunity to explore the three-dimensional solar atmosphere almost in real-time.
The telescope in Hawaii, equipped with specially developed instruments, can measure the sun’s magnetic field using polarized light and provides high-resolution images. Researchers use neural networks to evaluate the physical properties of the solar photosphere. To train the AI models, Yang’s team created a comprehensive dataset with simulated solar observations. A supercomputer generated 120 terabytes of data that mimic the telescope’s observations in extremely high resolution.
Out of this data, 13.7 terabytes have been made publicly available, along with detailed instructions. The researchers plan to release the deep learning models as a community tool, allowing amateur researchers to analyze solar observation data as well.
This effort is part of a broader initiative to use artificial intelligence to enhance our understanding of the sun and its impact on Earth. By making these tools available to the public, the researchers hope to foster a collaborative environment where both professionals and amateurs can contribute to solar research.
The study detailing these findings was published in The Astrophysical Journal. The combination of advanced solar observation and AI represents a significant step forward in predicting and understanding solar storms, which is crucial for mitigating their potential impact on Earth.
By using AI to process and analyze the vast amounts of data collected from solar observations, scientists aim to gain insights into the mechanisms that drive solar activity. This knowledge could eventually lead to better forecasting models, helping to protect technological infrastructure on Earth from the effects of solar storms.
As technology continues to evolve, the integration of AI in solar research exemplifies how interdisciplinary collaboration can lead to breakthroughs in scientific understanding. The ongoing development and refinement of AI models for solar observation highlight the potential for these tools to revolutionize the field, offering new ways to study and predict solar phenomena.