New science on experimenting with different machine-learning methods trained on white-light coronagraph data sets (SOHO/LASCO) of close-to-Sun CMEs, to estimate whether such a newly erupting ejection has the potential to induce geomagnetic activity has been published in the Astrophysical Journal.
Coronal mass ejections (CMEs) are the most geoeffective space weather phenomena. They are associated with large geomagnetic storms, and have the potential to cause disturbances to telecommunications, satellite network disruptions, and power grid damage and failures.
The paper focuses on forecasting the geoeffectiveness of CMEs to exclusively use solar onset parameters, in order to ensure extended warning times. The model proposed correctly identified 80.28% of the geoeffective events, with a precision of 15.19%, based on fivefold cross-validation on the entire data set.
The paper can be read here. Contact diana.ionescu (at) astro.ro.
Published on: Aug 06, 2022