Investigating and comparing the two superstorms in the 23rd solar cycle

• Published in: Indian journal of physics Vol. 96; no. 10; pp. 2707 - 2716
• Main Author: Basciftci, Fuat
• Format: Journal Article
• Language: English
• Published: New Delhi Springer India 01.08.2022; Springer Nature B.V
• Subjects: Artificial neural networks; Astrophysics and Astroparticles; Correlation analysis; Error analysis; Geomagnetism; Magnetic storms; Original Paper; Physics; Physics and Astronomy; Root-mean-square errors; Solar cycle; Solar wind; Artificial neural network (ANN); Geomagnetic storm; Solar wind parameters (SWp); Zonal geomagnetic indices (ZGi)
• ISSN: 0973-1458; 0974-9845
• DOI: 10.1007/s12648-022-02396-y

The present study investigated two super geomagnetic storms (GSs) that occurred on November 20, 2003 (Dst = –422 nT), and November 08, 2004 (Dst = –374 nT). The investigation was conducted by the zonal geomagnetic indices (ZGi) and the solar wind parameters (SWp) (nT). The analysis of the two super geomagnetic activities based on the artificial neural network model (ANNm) consists of the correlation matrix, the hierarchical cluster appearance, and the ZGi estimation with their errors. In the light of physical principles, the ANNm utilizes SWp as inputs and ZGi as outputs. The comparative estimation conclusions with a high correlation rate and low absolute error are satisfied. The ANNm performance is evaluated using the correlation constant ( R ) and root mean square error (RMSE) (nT) measures for either GS. The study claims that the network model is reliable, with an R -value of up to 88% and an RMSE value of down to 4.78 nT. The paper may support the Sun–Earth investigations.