-
- SM23D-2489: Plasma Wave Identification Near Shallow Plasmapause Gradients Using Unsupervised Machine Learning
-
Board 2489‚ Hall EFG (Poster Hall)NOLA CC
Author(s):Generic 'disconnected' Message
Samuel Buckler, University of Colorado Boulder (First Author, Presenting Author)
David Malaspina, University of Colorado
David Hartley, University of Iowa
Evan Tyler, University of Minnesota Twin Cities
Scott Thaller, University of Colorado Boulder
Jean-Francois Ripoll, CEA/DAM- ILE DE FRANCE
'Hiss' and 'chorus' are names for two types of plasma waves that play a critical role in loss and acceleration of electrons in Earth's magnetic field. Hiss is typically observed in a high plasma density region called the plasmasphere while chorus often exists outside this region.
Prior studies examined the properties of these waves, processing observations into wave models that serve as inputs to predictive simulations of Earth's radiation belts. However, the majority of these prior studies select for hiss and chorus waves detected near sharp plasma density gradients. Recent studies show that defining the plasmasphere boundary with sharp gradients excludes approximately half of all observation times. This bias may skew the wave models built from observational data.
In this study, a method of separating chorus and hiss is explored which does not use plasma density. Instead, chorus and hiss are separated using differences in the spread of their wave power. This information is used as input to a machine learning algorithm which groups different wave types. This enables a new quantification of hiss and chorus wave properties in regions where they mix, which have historically been excluded from past data analysis and models.
Scientific DisciplineNeighborhoodType
Enter Note
Go to previous page in this tab
Session
