Visualizing and Interpreting Unsupervised Solar Wind Classifications

Abstract

One of the goals of machine learning is to eliminate tedious and arduous repetitive work. The manual and semi-automatic classification of millions of hours of solar wind data from multiple missions can be replaced by automatic algorithms that can discover, in mountains of multi-dimensional data, the real differences in the solar wind properties. In this paper we present how unsupervised clustering techniques can be used to segregate different types of solar wind. We propose the use of advanced data reduction methods to pre-process the data, and we introduce the use of Self-Organizing Maps to visualize and interpret 14 years of ACE data. Finally, we show how these techniques can potentially be used to uncover hidden information, and how they compare with previous manual and automatic categorizations.

Submitted to: Frontiers in Astronomy and Space Sciences, Space Physics, Research Topic Machine Learning in Heliophysics

References in corpus (9)

• Kappa distributions: theory and applications in space plasmas
• Origins of the slow and the ubiquitous fast solar wind
• Enhanced Energy Transfer Rate in Solar Wind Turbulence Observed near the Sun from Parker Solar Probe
• Classification of Solar Wind with Machine Learning
• Magnification Control in Self-Organizing Maps and Neural Gas
• Turbulence Transport Modeling and First Orbit Parker Solar Probe (PSP) Observations
• Coronal sources and in situ properties of the solar winds sampled by ACE during 1999-2008
• The nature of Elsässer variables in compressible MHD
• Characterizing magnetic reconnection regions using Gaussian mixture models on particle velocity distributions

Cited by in corpus (4)

• A review of unsupervised learning in astronomy
• Unsupervised classification of simulated magnetospheric regions
• Scope and limitations of ad hoc neural network reconstructions of solar wind parameters
• Clustering Wind data at 1 AU to contextualize magnetic reconnection in the solar wind