|Solar flares can reach velocities of up to several million kilometers per hour. |
Illustration: Matti Ahlgren/Aalto University
Associate Professor Maarit Korpi-Lagg has received funding from the European Research Council to develop a forecasting tool to locate the source regions for the eruption of solar flares already a few days before they emerge on the Sun’s surface.
The Earth is constantly bombarded by a stream of particles from the Sun, called solar wind. This stream can escalate into storms, which are born from massive solar flares spewing out from the Sun’s highly magnetized active regions. When strong solar storms hit Earth, they can have massive repercussions for telecommunications, global positioning systems and electrical grids.
In July 2012, the most severe solar flare in 150 years was spat out by the Sun. Fortunately, the resulting solar storm missed Earth. Had it been directed toward us; it would have had the potential to leave societies and the global economy in shatters and taken years to recover from.
‘Only the worst solar storms are a real threat to human life. However, the costs of fixing damages and shielding our digitalized infrastructure from them, are very high,’ says Maarit Korpi-Lagg, associate professor at Aalto University.
|‘The forecasting tools we’ll be developing in the new project will make our societies less vulnerable to bad space weather and help save resources in building damage-prevention capacity,’ says Korpi-Lagg.|
Image: Matti Ahlgren/Aalto University
Korpi-Lagg is developing tools for short and long-term forecasting of solar active regions with her astroinformatics research group. They have just received the Proof-of-Concept grant from the European Research Council (ERC) for their Solar cycle prediction tool using solar internal oscillations (SYCOS) project. The funding is given to research that aims to harness basic research to be used for the benefit of society.
‘The forecasting tools we’ll be developing in the new project will make our societies less vulnerable to bad space weather and help save resources in building damage-prevention capacity,’ says Korpi-Lagg.
Preparing for the worst
Solar activity runs in 11-year cycles and the next peak is estimated to be around 2025. Modern methods focus on monitoring the Sun’s surface activity, giving little lead time to prepare for solar storms, which can travel through space at extremely high velocities.
‘Currently, solar flares and adverse space weather can only be evaluated after active areas emerge on the Sun’s surface. At this point, it might be too late to mount an efficient response or manage risks on Earth,’ says Korpi-Lagg.
Her research group has developed an alternative forecasting method by focusing on the location and properties of the magnetic fields below the Sun’s surface.
‘Observing the location and topology of sub-surface magnetic fields helps us predict the formation of active regions a few days prior to current methods. This method will greatly enhance the predictability and preparedness for fast-moving space weather phenomena, potentially enroute to the Earth’s magnetosphere,’ says Korpi-Lagg.
The fresh funding will enable the team to develop the open-source forecasting tools and to recruit more data analysis and machine learning expertise. Korpi-Lagg’s research group is working in close cooperation with the Max Planck Institute for Solar System Research (MPS) in the new project. The project leans heavily on the basic research project, in which the research group, together with MPS, developed a novel method to observe sub-surface magnetic fields indirectly – a feat that wasn’t possible before.
Putting solar data in its place
Solar data is vast and complex. To analyze and understand the data, researchers need to complement their models and knowledge with high-performance computing capacity as well as machine learning.
‘The tools we’re developing in the new project can orchestrate the whole forecasting process from start to finish: oscillation data – meaning the way the Sun’s internal waves fluctuate – is gathered from databases around the world, analyzed in real-time, and then our machine learning model is used to create a prediction of the Sun’s magnetic weather,’ Korpi-Lagg explains.
The research group’s aim is to integrate the open-source tool into existing forecasting systems in space weather stations to enhance our capabilities to mitigate the risks of bad space weather.
Source/Credit: Aalto University