Solar storms are usually weak enough that Earth’s atmosphere and magnetic field can sufficiently protect the planet; However, in the case of large-scale solar storms, the impact can be much larger and range from high voltages in the power grid, disrupted GPS and aircraft radio connections, to nationwide internet failures and widespread power outages.
Artificial intelligence should help
However, the better the forecast, the easier it will be to respond in time and minimize losses. In Graz, data experts from the No-Center are now working with researchers from the Austrian Academy of Sciences’ Institute for Space Research (IWF) to develop a prediction tool based on artificial intelligence. They are part of the EU project “Europlanet 2024 – Research Infrastructure”, which aims to more strongly promote European research in the field of planetary science.
A solar storm does not equal a solar storm
According to Graz researchers, the Sun constantly sends radiation and charged particles into space – the solar wind. However, so-called coronal mass ejections, commonly known as solar storms, sometimes cause a rapid increase in particle density.
The ability of solar storms to generate extreme geomagnetic storms, in turn, depends largely on the orientation of their magnetic field – in technical jargon this is known as the “BZ magnetic field component”. Here, the relative orientation of this component to Earth’s magnetic field determines how much energy is transferred to Earth’s magnetic field: the stronger the Bz component is to the south, the greater the risk of a massive geomagnetic storm. Will happen.
So far, the Bz magnetic field component cannot be predicted with sufficient advance warning before a solar storm hits Earth. However, machine learning allows researchers to train algorithms to analyze large amounts of data and derive predictions and new solutions from it.
“Data measured by the probe directly into the solar wind takes only minutes to transmit to Earth. We first looked to see if the information about the first few hours of a solar storm is sufficient to be able to predict its strength,” Hannah Rudisser from No-Center explains. “The European research network ‘EuroPlanet 2024’ contains a vast wealth of data from space missions, simulations and laboratory experiments. We aim to extract the knowledge contained in this data and make it usable.”
Data from about 350 different solar storms
To “train” the prediction tool, the researchers fed the program with data from about 350 different solar storms since 2007. Now, to test the forecasting tool in experimental real-time mode, the team simulates how solar storms are measured by spacecraft and evaluates how the continuous feeding of new information improves predictions. “Our forecasting tool can predict the Bz component very well. It works especially well when we use data from the first four hours of a solar storm’s magnetic core,” Rudisser summarizes the first results. Is.
automatic as target
It is hoped that the new space missions will provide additional data that will further increase the accuracy of the predictions. Ultimately, the researchers would like to use AI methods to automatically detect solar storms in the solar wind – this will enable them to continuously identify solar storms without a human user to be able to use the method in real time. required to be enabled.
Freelance twitter maven. Infuriatingly humble coffee aficionado. Amateur gamer. Typical beer fan. Avid music scholar. Alcohol nerd.