Mathematicians at the University of Sheffield have explored the reliability and limitations of the state-of-the-art artificial intelligence techniques in predicting the solar magnetic field.
The new research, published in Nature Astronomy, addresses the critical question of what limitations AI models have in predicting the solar magnetic field, which could provide new insight into space weather forecasting.
Space weather refers to the physical conditions in the space near the Earth, which determines the functioning and safety of a wide range of modern high-tech based facilities including airplanes, satellites, GPS, power grids and oil pipelines.
A number of novel AI techniques have been used for a variety of purposes in the area of solar physics and space weather forecasting. An advanced AI model has been proposed to generate the photospheric magnetic fields observed by NASA’s SDO satellite mission, using observations of the solar chromosphere with temperatures around 50,000 degrees as the only inputs.
Promising results have been achieved with a high level of cross-correlation between the total unsigned magnetic flux of the observed and the generated magnetic fields of the visible surface of the Sun, called the solar photosphere.
The photospheric magnetic fields of our star the Sun play a fundamental role in forming sunspots, the active regions in the solar atmosphere. Three regions are also believed to host the onset of coronal mass ejections (CMEs) and one of the largest and most energetic explosions in our Solar System, the flares.
These gigantic eruptive features are the most dominant drivers of severe space weather events.
Professor Robert Erdélyi, head of the Solar Physics and Space Plasma Research Centre (SP2RC) in the University of Sheffield’s School of Mathematics and Statistics, said: “Mathematics and physics are the basics, it is always somewhat dangerous to ignore the mathematical modelling of the physics behind these solar activities when trying to predict them using techniques such as AI. We must not use these fascinating tools as black boxes as this may mislead us and one may arrive at over-interpreted conclusions.”
Dr JiaJia Liu, from the University of Sheffield, said: “Though the AI model performed well in predicting the total unsigned magnetic flux, its performance was not as expected when predicting other parameters including net flux and polarity reversion lines, which are vital in space weather predictions.”
Professor Erdélyi added: “AI is an advanced and rapidly developing technique that has wide applications in science and technology. However, users must be aware of the limitations, especially when applying it to scientific research. Without considering the fundamental mathematics and physics associated, this technique is limited to achieve reliable results even if one is using the most advanced AI or machine Learning.”
The full team of researchers includes Dr Jiajia Liu, Dr Yimin Wang, Dr Ye Jiang and Professor Robert Erdélyi (University of Sheffield), Dr Xin Huang (Chinese Academy of Science), Dr Marianna B. Korsos (Aberystwyth University), and Professor Yuming Wang (University of Science and Technology of China).