Classification of Solar Wind with Machine Learning

Abstract

We present a four-category classification algorithm for the solar wind, based on Gaussian Process. The four categories are the ones previously adopted in Xu & Borovsky [2015]: ejecta, coronal hole origin plasma, streamer belt origin plasma, and sector reversal origin plasma. The algorithm is trained and tested on a labeled portion of the OMNI dataset. It uses seven inputs: the solar wind speed $V_{sw}$, the temperature standard deviation ${\sigma }_T$, the sunspot number $R$, the $f_{10.7}$ index, the Alfven speed $v_A$, the proton specific entropy $S_p$ and the proton temperature $T_p$ compared to a velocity-dependent expected temperature. The output of the Gaussian Process classifier is a four element vector containing the probabilities that an event (one reading from the hourly-averaged OMNI database) belongs to each category. The probabilistic nature of the prediction allows for a more informative and flexible interpretation of the results, for instance being able to classify events as 'undecided'. The new method has a median accuracy larger than $90\%$ for all categories, even using a small set of data for training. The Receiver Operating Characteristic curve and the reliability diagram also demonstrate the excellent quality of this new method. Finally, we use the algorithm to classify a large portion of the OMNI dataset, and we present for the first time transition probabilities between different solar wind categories. Such probabilities represent the 'climatological' statistics that determine the solar wind baseline.

accepted in J. Geophys. Res

References in corpus (2)

• Properties and Geoeffectiveness of Magnetic Clouds during Solar Cycles 23 and 24
• Sources of The Slow Solar wind During the Solar Cycle 23/24 Minimum

Cited by in corpus (18)

• The multi-scale nature of the solar wind
• The Challenge of Machine Learning in Space Weather Nowcasting and Forecasting
• The data-driven future of high energy density physics
• Radial evolution of the solar wind in pure high-speed streams: HELIOS revised observations
• Automatic classification of plasma regions in near-Earth space with supervised machine learning: application to Magnetospheric Multi Scale 2016-2019 observations
• Helium abundance and speed difference between helium ions and protons in the solar wind from coronal holes, active regions, and quiet Sun
• Automatic detection of Interplanetary Coronal Mass Ejections from in-situ data: a deep learning approach
• A New Automatic Tool for CME Detection and Tracking with Machine Learning Techniques
• Revealing an unexpectedly low electron injection threshold via reinforced shock acceleration
• Beyond optimization -- supervised learning applications in relativistic laser-plasma experiments
• Visualizing and Interpreting Unsupervised Solar Wind Classifications
• Estimating uncertainties in the back-mapping of the fast solar wind
• Exploring the potential of neural networks to predict statistics of solar wind turbulence
• Proton-proton collisional age to order solar wind types
• Which Upstream Solar Wind Conditions Matter Most in Predicting Bz within Coronal Mass Ejections
• Clustering Wind data at 1 AU to contextualize magnetic reconnection in the solar wind
• Machine Learning Approach for Solar Wind Categorization
• Scope and limitations of ad hoc neural network reconstructions of solar wind parameters