Coronal mass ejections (CMEs) are massive solar eruptions of magnetized plasma . When directed toward Earth, they interact with Earth’s magnetosphere, which acts as a shield against solar radiation, temporarily weakening its ability to deflect ionizing particles carried by the CME. These then flow along the outer edge of the magnetosphere, in the direction of Earth’s magnetic poles, where the shield is weakest, making contact with Earth’s atmosphere. Their interaction gives rise to polar auroras, spectacular dynamic phenomena of lights, shapes and colors, usually visible at high latitudes.

Figure 1: Aurora borealis observed Sunday, November 5, from several sites in Italy.

The impact of a CME on Earth, however, is also associated with potentially damaging events for orbiting satellites, telecommunications, navigation systems, and ground facilities (such as power grids: a well-known example is the blackout that hit Quebec in March 1989). These are associated with, in the most extreme cases, possible biological hazards to astronauts.

The accurate and timely prediction of the arrival of a CME to Earth as well as the level of induced geomagnetic storm has thus become, especially recently, an extremely hot topic that has attracted the attention of many research institutes aiming to provide forecast models for these potentially dangerous solar events. In particular, the European Space Agency (ESA) has set up the Space Weather Service Network (SWESNET), designed to collect all the contributions provided by the different European research institutes. In July this year, the INAF-Turin Astrophysical Observatory provided ESA with an innovative forecasting algorithm (named Geomagnetic Effectiveness tool) based on the analysis of in-situ measurements of CMEs acquired by spacecraft orbiting around the Lagrangian point L1, located 1.5 million km from Earth in the Sunward direction. This is the first one on behalf of the National Institute of Astrophysics to be integrated onto the ESA Space Weather Service Network. The tool, developed by the INAF-Turin Astrophysical Observatory, as part of the “Heliospheric Space Weather Initiative” - HelioMeteo for short - which brings together collaborations with Aerospace Logistics Technology Engineering Company (ALTEC) and the University of Genoa, allowed accurate detection of the November 5 CME about 9 hours before the onset of the geomagnetic storm.

Figure 2: Output of the Geomagnetic Effectiveness tool: the CME, identified by the blue arrow, was revealed 9 hours ahead of the onset of the geomagnetic storm (red and blue dashed vertical lines, respectively).

“This is the first prediction of a space weather event made by an operational forecasting software developed by INAF for the Italian contribution to the ESA Space Weather Service Network”, says Silvano Fineschi of the INAF-Turin Astrophysical Observatory and head of the “HelioMeteo” initiative.

“The operational handling of the products Geomagnetic Effectiveness tool and CME Propagation Prediction tool is a matter of responsibility for ALTEC. For validation activities, the integration of scientific algorithms, including those exploiting machine learning techniques, and the integration of the products onto the ESA-SWESNET portal, the infrastructure of the Heliospheric Space Weather Data Center, developed and maintained by ALTEC at its Turin site, is used”, adds Filomena Solitro, program manager of the ALTEC “Science and Advanced Data Processing” sector.This result demonstrates not only the efficiency of the tool in predicting the imminent impact of CMEs on Earth, but also how the use of in-situ data can be successful in detecting and forecasting their potential geo-effectiveness. The INAF-Turin Astrophysical Observatory research group is currently working on the development and prototyping of other promising forecasting tools and models. In particular, it will be the lead partner, for INAF, of the Spoke Space, focused on Space Weather, in the PNRR SPACE IT UP program. In addition, it has over the years established a fruitful collaboration in HelioMeteo with the Department of Mathematics (DIMA) of the University of Genoa, for example in the context of the AIxtreme project, coordinated by Prof. Anna Maria Massone. The group, led by Prof. Michele Piana, will employ the use of Artificial Intelligence (AI) in the development of new and even more accurate forecasting algorithms, exploring the full potential of AI in what is regarded as the approach of the future in space weather science. “This innovative strategy for Space Weather prediction” - says Michele Piana - “is based on the design of neural networks that are trained using archives of past and current space missions and are then able to forecast all types of events that characterize Space Weather, from solar flares, through the time of flight of CMEs, to magnetosphere impact measurements”.