Abstract:Collisionless space plasma environments are typically characterized by distinct particle populations. Although moments of their velocity distribution functions help in distinguishing different plasma regimes, the distribution functions themselves provide more comprehensive information about the plasma state, especially at times when the distribution function includes non-thermal effects. Unlike moments, however, distribution functions are not easily characterized by a small number of parameters, making their classification more difficult to achieve. In order to perform this classification, we propose to distinguish between the different plasma regions by applying dimensionality reduction and clustering methods to electron distributions in pitch angle and energy space. We utilize four separate algorithms to achieve our plasma classifications: autoencoders, principal component analysis, mean shift, and agglomerative clustering. We test our classification algorithms by applying our scheme to data from the Cluster-Plasma Electron and Current Experiment instrument measured in the Earth’s magnetotail. Traditionally, it is thought that the Earth’s magnetotail is split into three different regions (the plasma sheet, the plasma sheet boundary layer, and the lobes), that are primarily defined by their plasma characteristics. Starting with the ECLAT database with associated classifications based on the plasma parameters, we identify eight distinct groups of distributions, that are dependent upon significantly more complex plasma and field dynamics. By comparing the average distributions as well as the plasma and magnetic field parameters for each region, we relate several of the groups to different plasma sheet populations, and the rest we attribute to the plasma sheet boundary layer and the lobes. We find clear distinctions between each of our classified regions and the ECLAT results. The automated classification of different regions in space plasma environments provides a useful tool to identify the physical processes governing particle populations in near-Earth space. These tools are model independent, providing reproducible results without requiring the placement of arbitrary thresholds, limits or expert judgment. Similar methods could be used onboard spacecraft to reduce the dimensionality of distributions in order to optimize data collection and downlink resources in future missions.

Classifying Different Types of Solar-Wind Plasma with Uncertainty Estimations Using Machine Learning

Classifying different types of solar wind plasma with uncertainty estimations using machine learning

Machine Learning Approach for Solar Wind Categorization

Objectively Determining States of the Solar Wind Using Machine Learning

Using Dimensionality Reduction and Clustering Techniques to Classify Space Plasma Regimes

Forecasting Geoffective Events from Solar Wind Data and Evaluating the Most Predictive Features through Machine Learning Approaches

Investigation of the relationship between geomagnetic activity and solar wind parameters based on a novel neural network (potential learning)

Solar wind prediction using deep learning

Combining Empirical and Physics-Based Models for Solar Wind Prediction

Improvement of Plasma Sheet Neural Network Accuracy With Inclusion of Physical Information

Visualizing and Interpreting Unsupervised Solar Wind Classifications

Machine learning model of the plasmasphere to forecast satellite charging caused by solar storms.

Understanding and Modeling the Dynamics of Storm-time Atmospheric Neutral Density using Random Forests

Challenges and the next transformative steps in understanding plasma turbulence from the perspective of multi-spacecraft measurements

The Effect of Fast Solar Wind on Ion Distribution Downstream of Earth's Bow Shock

Physics-driven machine learning for the prediction of coronal mass ejections' travel times

Solar wind data analysis aided by synthetic modeling: a better understanding of plasma-frame variations from temporal data

Forecasting High-Speed Solar Wind Streams from Solar Images

Using gradient boosting regression to improve ambient solar wind model predictions

Identification of Flux Rope Orientation via Neural Networks

Using Recurrent Neural Networks to improve initial conditions for a solar wind forecasting model