engleski

Empirical statistical model for predicting geomagnetic storm levels based on remote solar observations

One of main issues of space weather is prediction of strongly geoeffective events. Monitoring near-Earth interplanetary parameters can give a quite reliable prediction of the potentially harmful events, however the warnings precede the event only about 1 hour in advance (for monitoring spacecrafts at L1), providing very limited amount of reaction time. For early warnings one should try to employ remote solar observations instead. However, with current knowledge on the coronal mass ejections (CMEs), main drivers of largest geomagnetic storms, we are still not able to predict the arrival time, velocity and magnetic field, or even if it will entirely miss the Earth. Therefore, an empirical statistical model was established that can be used as an early geomagnetic storm warning. For that purpose 211 front-sided, solar flare-associated CMEs with a minimal speed of 400 km/sec have been selected from the LASCO/SOHO catalog. Furthermore, association with a specific Dst (disturbance storm time) index was made. The resulting sample contains geoeffective and non-geoeffective CMEs, as well as CMEs that missed the Earth. An extensive statistical analysis was performed to determine the probability distributions for Dst depending on the CME and flare characteristics. Several CME and flare parameters were investigated, as well as the effect of successive CMEs occurrence to the change in the probability of certain Dst index values. The results confirmed some previously known connections between remote solar properties and geomagnetic storms, namely the importance of CME speed, apparent width, source position and associated solar flare type. For the first time we quantify these relationships and use them to construct a statistical model for predicting the probability of geomagnetic storm level based on remote solar observations of CMEs and flares. This work has received funding from the European Commission FP7 Project COMESEP (263252).

space weather ; coronal mass ejections ; geomagnetic storms

nije evidentirano