Automated Classification of Plasma Regions Using 3D Particle Energy Distributions

@article{Olshevsky2021AutomatedCO,
  title={Automated Classification of Plasma Regions Using 3D Particle Energy Distributions},
  author={Vyacheslav Olshevsky and Yu. V. Khotyaintsev and Andrey Divin and Gian Luca Delzanno and Sven Anderzen and Pawel Herman and Steven W. D. Chien and Levon A. Avanov and Stefano Markidis},
  journal={Journal of Geophysical Research: Space Physics},
  year={2021},
  volume={126}
}
We investigate the properties of the ion sky maps produced by the Dual Ion Spectrometers (DIS) from the Fast Plasma Investigation (FPI). We have trained a convolutional neural network classifier to predict four regions crossed by the Magnetospheric Multiscale Mission (MMS) on the dayside magnetosphere: solar wind, ion foreshock, magnetosheath, and magnetopause using solely DIS spectrograms. The accuracy of the classifier is >98 %. We use the classifier to detect mixed plasma regions, in… 

Automatic Classification of Plasma Regions in Near-Earth Space With Supervised Machine Learning: Application to Magnetospheric Multi Scale 2016–2019 Observations

This study applies the fully convolutional neural network (FCN) deep machine-leaning algorithm to the recent Magnetospheric Multi Scale mission data in order to classify 10 key plasma regions in near-Earth space for the period 2016-2019, indicating that such method can be successfully applied to any in situ spacecraft plasma database.

Automatic detection of the Earth Bow Shock and Magnetopause from in-situ data with machine learning

The method was used to identify 14186 magnetopause crossings and 16192 bow shock crossings in the data of 10 different spacecrafts of the THEMIS, ARTEMIS, Cluster and Double Star missions and for a total of 79 cumulated years.

A UTOMATIC CLASSIFICATION OF PLASMA REGIONS IN NEAR-E ARTH SPACE WITH SUPERVISED MACHINE LEARNING : APPLICATION

  • Computer Science
  • 2020
This study applies the fully convolutional neural network (FCN) deep machine-leaning algorithm to the recent Magnetospheric Multi Scale mission data in order to classify ten key plasma regions in near-Earth space for the period 2016-2019, indicating that the FCN method is reliable to accurately classify labeled time series data.

Classification of Cassini’s Orbit Regions as Magnetosphere, Magnetosheath, and Solar Wind via Machine Learning

Several machine learning algorithms and feature subsets from a variety of particle and magnetic field instruments on-board the Cassini spacecraft were explored for their utility in classifying orbit

Unsupervised classification of simulated magnetospheric regions

An unsupervised classification method for magnetospheric regions, that could constitute the first-step of a multi-step method for the automatic identification of magnetosphere processes of interest, based on Self Organizing Maps.

Automated characterization of magnetic reconnection using particle distributions

Magnetic reconnection is a fundamental process for many plasma phenomena converting the stored magnetic energy into kinetic energy, thermal energy, and particle acceleration energy. Various missions

Massive Multi‐Mission Statistical Study and Analytical Modeling of the Earth's Magnetopause: 1. A Gradient Boosting Based Automatic Detection of Near‐Earth Regions

We present an automatic classification method of the three near‐Earth regions, the magnetosphere, the magnetosheath and the solar wind from their in situ data measurement by multiple spacecraft.

Classifying the Magnetosheath Behind the Quasi‐Parallel and Quasi‐Perpendicular Bow Shock by Local Measurements

We investigate and evaluate the possibility of using local magnetosheath measurements to classify the plasma according to upstream conditions. In order to do this, we use simultaneous measurements

A Database of MMS Bow Shock Crossings Compiled Using Machine Learning

A machine learning approach is used to automatically identify shock crossings from the Magnetospheric Multiscale (MMS) spacecraft, and it is found that quasi‐parallel shocks are more efficient at accelerating ions than quasi‐perpendicular shocks.

MMS SITL Ground Loop: Automating the Burst Data Selection Process

A Long-Short Term Memory (LSTM) Recurrent Neural Network model is introduced to detect MP crossings and assist studies of energy transfer into the magnetosphere and its predictions are public to ease the burden of arduous event searches involving the MP.

References

SHOWING 1-10 OF 38 REFERENCES

Automatic Classification of Plasma Regions in Near-Earth Space With Supervised Machine Learning: Application to Magnetospheric Multi Scale 2016–2019 Observations

This study applies the fully convolutional neural network (FCN) deep machine-leaning algorithm to the recent Magnetospheric Multi Scale mission data in order to classify 10 key plasma regions in near-Earth space for the period 2016-2019, indicating that such method can be successfully applied to any in situ spacecraft plasma database.

Large-Scale Statistical Survey of Magnetopause Reconnection

This work introduced a hierarchical Bayesian mixture model with linear and auto regressive components to identify the magnetopause, which will allow scientists to study the micro-physics of reconnection events in the Magnetospheric Multiscale Mission using the large body of MMS data without manual classification.

MMS SITL Ground Loop: Automating the Burst Data Selection Process

A Long-Short Term Memory (LSTM) Recurrent Neural Network model is introduced to detect MP crossings and assist studies of energy transfer into the magnetosphere and its predictions are public to ease the burden of arduous event searches involving the MP.

Fast Plasma Investigation for Magnetospheric Multiscale

The Fast Plasma Investigation (FPI) was developed for flight on the Magnetospheric Multiscale (MMS) mission to measure the differential directional flux of magnetospheric electrons and ions with

Magnetospheric Multiscale Science Mission Profile and Operations

The Magnetospheric Multiscale (MMS) mission and operations are designed to provide the maximum reconnection science. The mission phases are chosen to investigate reconnection at the dayside

Magnetospheric Multiscale Overview and Science Objectives

Magnetospheric Multiscale (MMS), a NASA four-spacecraft constellation mission launched on March 12, 2015, will investigate magnetic reconnection in the boundary regions of the Earth’s magnetosphere,

Large‐scale characteristics of reconnection diffusion regions and associated magnetopause crossings observed by MMS

The Magnetospheric Multiscale (MMS) mission was designed to make observations in the very small electron diffusion region (EDR), where magnetic reconnection takes place. From a data set of over 4500

The Magnetospheric Multiscale Magnetometers

To achieve mission objectives, the calibration determined on the ground will be refined in space to ensure all eight magnetometers are precisely inter-calibrated, and the information flow that provides the data on the rapid time scale needed for mission success is described.

Magnetospheric Multiscale Instrument Suite Operations and Data System

The MMS science operations approach, processes and data systems, including the burst system and the SITL concept are described, which are designed to maximize the value of the burst data by prioritizing the data segments selected for transmission to the ground.

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

The recurring themes throughout the review are the need to shift the authors' forecasting paradigm to a probabilistic approach focused on the reliable assessment of uncertainties, and the combination of physics‐based and machine learning approaches, known as gray box.