Tracking Equatorial Plasma Bubbles Based on an Artificial Neural Network Classifier

Once an equatorial plasma bubble has been identified, tracking the same bubble as it evolves in time and position between measurements helps determine its impact on communications and drift velocity. For communication, older bubbles have less of an impact than newly formed ones. Drift velocity is often poorly represented in existing models, therefore re-identification and associating additional measurements to a previously measured bubble will improve our understanding of how the drift velocity changes over a bubbles lifetime. We present a brief history of how machine learning has been used in the field of Space Weather along with our study on applying an artificial neural network classifier to associate measurements to a previously identified plasma bubble.