Utilizing ionospheric Total Electron Content (TEC) data for Solar flare predictions using Support Vector Machine (SVM)
Abstract
Utilizing ground and space-based remote sensing technology is essential for near-space environment studies and space weather research. Forecasting where and when space weather events such as solar flares and X-rays bursts are more, or less, likely to occur in a certain area of interest constitute a significant challenge in space weather research. Although most of the governing forces and basic processes of space weather events can be modeled numerically, in the absence of sufficiently detailed and real-time data, accurate forecasting of such events remains a challenging task. Space weather scientist are therefore gradually exploring multivariate data analysis techniques from the fields of data mining or machine learning in order to approximate future occurrences of space weather events from past distribution patterns. Here, we present the use of Support Vector Machin (SVM) applied with ionospheric Total Electron Content (TEC), derived from worldwide GPS geodetic receiver network, in order to predict B, C, M and X-class solar flare events. The results are shown as a confusion matrix up to three days before each tested class events. The proposed method has the ability to predict solar flare events with 80-95% accuracy, even with a small number of solar flares induced TEC maps.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMNH33C0930R
- Keywords:
-
- 4301 Atmospheric;
- NATURAL HAZARDS;
- 4302 Geological;
- NATURAL HAZARDS;
- 4305 Space weather;
- NATURAL HAZARDS;
- 4315 Monitoring;
- forecasting;
- prediction;
- NATURAL HAZARDS