A Prediction Model for the Geomagnetic Index Kp using Machine Learning Algorithms

The Sun is the primary driver of geomagnetic activity on Earth, as its solar wind and solar flares interact with the Earth's magnetosphere, causing fluctuations and disturbances that are measured by geomagnetic indices like the Kp index. High Kp values can lead to beautiful auroras but also pose risks to telecommunications infrastructure and human activities reliant on stable electromagnetic conditions. Consequently, a model predicting the Kp index is of significant interest. Using Machine learning algorithms, a model predicting Kp index to an accuracy greater than 80% using data from Low-Resolution OMNI (LRO) dataset¹ will be presented. Correlation with the daily total sunspot number will also be presented. In addition, several machine learning and deep learning algorithms applicable to the data will be compared.





[1] King, J. H., and N. E. Papitashvili (2005), Solar wind spatial scales in and comparisons of hourly Wind and ACE plasma and magnetic field data, J. Geophys. Res., 110, A02104 (2005)