A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind
The Wind spacecraft has yielded several decades of high-resolution magnetic field data, a large fraction of which displays small-scale structures. In particular, the solar wind is full of wavelike fluctuations that appear in both the field magnitude and its components. The nature of these fluctuatio...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IOP Publishing
2023-01-01
|
Series: | The Astrophysical Journal |
Subjects: | |
Online Access: | https://doi.org/10.3847/1538-4357/acc8d5 |
_version_ | 1797696207284338688 |
---|---|
author | Samuel Fordin Michael Shay Lynn B. Wilson III Bennett Maruca Barbara J. Thompson |
author_facet | Samuel Fordin Michael Shay Lynn B. Wilson III Bennett Maruca Barbara J. Thompson |
author_sort | Samuel Fordin |
collection | DOAJ |
description | The Wind spacecraft has yielded several decades of high-resolution magnetic field data, a large fraction of which displays small-scale structures. In particular, the solar wind is full of wavelike fluctuations that appear in both the field magnitude and its components. The nature of these fluctuations can be tied to the properties of other structures in the solar wind, such as shocks, that have implications for the time evolution of the solar wind. As such, having a large collection of wave events would facilitate further study of the effects that these fluctuations have on solar wind evolution. Given the large volume of magnetic field data available, machine learning is the most practical approach to classifying the myriad small-scale structures observed. To this end, a subset of Wind data is labeled and used as a training set for a multibranch 1D convolutional neural network aimed at classifying circularly polarized wave modes. Using this algorithm, a preliminary statistical study of 1 yr of data is performed, yielding about 300,000 wave intervals out of about 5,000,000 solar wind intervals. The wave intervals come about more often in the fast solar wind and at higher temperatures, and the number of waves per day is highly periodic. This machine learning–based approach to wave detection has the potential to be a powerful, inexpensive way to catalog waves throughout decades of spacecraft data. |
first_indexed | 2024-03-12T03:22:23Z |
format | Article |
id | doaj.art-856802dd0f3e4230bc70c58c65ce6c09 |
institution | Directory Open Access Journal |
issn | 1538-4357 |
language | English |
last_indexed | 2024-03-12T03:22:23Z |
publishDate | 2023-01-01 |
publisher | IOP Publishing |
record_format | Article |
series | The Astrophysical Journal |
spelling | doaj.art-856802dd0f3e4230bc70c58c65ce6c092023-09-03T13:49:29ZengIOP PublishingThe Astrophysical Journal1538-43572023-01-0194924010.3847/1538-4357/acc8d5A Machine Learning–Based Approach to Time-series Wave Identification in the Solar WindSamuel Fordin0https://orcid.org/0000-0002-1634-9122Michael Shay1https://orcid.org/0000-0003-1861-4767Lynn B. Wilson III2https://orcid.org/0000-0002-4313-1970Bennett Maruca3https://orcid.org/0000-0002-2229-5618Barbara J. Thompson4https://orcid.org/0000-0001-6952-7343University of Delaware , 104 The Green, Newark, DE 19711, USA ; sfordin@udel.eduUniversity of Delaware , 104 The Green, Newark, DE 19711, USA ; sfordin@udel.eduNASA Goddard Space Flight Center , Greenbelt, MD 20771, USAUniversity of Delaware , 104 The Green, Newark, DE 19711, USA ; sfordin@udel.eduNASA Goddard Space Flight Center , Greenbelt, MD 20771, USAThe Wind spacecraft has yielded several decades of high-resolution magnetic field data, a large fraction of which displays small-scale structures. In particular, the solar wind is full of wavelike fluctuations that appear in both the field magnitude and its components. The nature of these fluctuations can be tied to the properties of other structures in the solar wind, such as shocks, that have implications for the time evolution of the solar wind. As such, having a large collection of wave events would facilitate further study of the effects that these fluctuations have on solar wind evolution. Given the large volume of magnetic field data available, machine learning is the most practical approach to classifying the myriad small-scale structures observed. To this end, a subset of Wind data is labeled and used as a training set for a multibranch 1D convolutional neural network aimed at classifying circularly polarized wave modes. Using this algorithm, a preliminary statistical study of 1 yr of data is performed, yielding about 300,000 wave intervals out of about 5,000,000 solar wind intervals. The wave intervals come about more often in the fast solar wind and at higher temperatures, and the number of waves per day is highly periodic. This machine learning–based approach to wave detection has the potential to be a powerful, inexpensive way to catalog waves throughout decades of spacecraft data.https://doi.org/10.3847/1538-4357/acc8d5Solar windInterplanetary physicsSpace weatherSpace plasmasNeural networksConvolutional neural networks |
spellingShingle | Samuel Fordin Michael Shay Lynn B. Wilson III Bennett Maruca Barbara J. Thompson A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind The Astrophysical Journal Solar wind Interplanetary physics Space weather Space plasmas Neural networks Convolutional neural networks |
title | A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind |
title_full | A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind |
title_fullStr | A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind |
title_full_unstemmed | A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind |
title_short | A Machine Learning–Based Approach to Time-series Wave Identification in the Solar Wind |
title_sort | machine learning based approach to time series wave identification in the solar wind |
topic | Solar wind Interplanetary physics Space weather Space plasmas Neural networks Convolutional neural networks |
url | https://doi.org/10.3847/1538-4357/acc8d5 |
work_keys_str_mv | AT samuelfordin amachinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT michaelshay amachinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT lynnbwilsoniii amachinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT bennettmaruca amachinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT barbarajthompson amachinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT samuelfordin machinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT michaelshay machinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT lynnbwilsoniii machinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT bennettmaruca machinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind AT barbarajthompson machinelearningbasedapproachtotimeserieswaveidentificationinthesolarwind |