Machine learning reveals systematic accumulation of electric current in lead-up to solar flares

Dattaraj B. Dhuri, Shravan Hanasoge, Mark C.M. Cheung

    Research output: Contribution to journalArticle

    Abstract

    Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.

    Original languageEnglish (US)
    Pages (from-to)11141-11146
    Number of pages6
    JournalProceedings of the National Academy of Sciences of the United States of America
    Volume166
    Issue number23
    DOIs
    StatePublished - Jun 4 2019

    Fingerprint

    Solar Activity
    Magnetic Fields
    Physics
    Weather
    Radiation
    Lead
    Machine Learning

    Keywords

    • Machine learning
    • Solar flares
    • Solar magnetic fields
    • Space weather

    ASJC Scopus subject areas

    • General

    Cite this

    Machine learning reveals systematic accumulation of electric current in lead-up to solar flares. / Dhuri, Dattaraj B.; Hanasoge, Shravan; Cheung, Mark C.M.

    In: Proceedings of the National Academy of Sciences of the United States of America, Vol. 166, No. 23, 04.06.2019, p. 11141-11146.

    Research output: Contribution to journalArticle

    @article{5f98c0b8a2ba4e259d0ce58d4b722dc9,
    title = "Machine learning reveals systematic accumulation of electric current in lead-up to solar flares",
    abstract = "Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.",
    keywords = "Machine learning, Solar flares, Solar magnetic fields, Space weather",
    author = "Dhuri, {Dattaraj B.} and Shravan Hanasoge and Cheung, {Mark C.M.}",
    year = "2019",
    month = "6",
    day = "4",
    doi = "10.1073/pnas.1820244116",
    language = "English (US)",
    volume = "166",
    pages = "11141--11146",
    journal = "Proceedings of the National Academy of Sciences of the United States of America",
    issn = "0027-8424",
    number = "23",

    }

    TY - JOUR

    T1 - Machine learning reveals systematic accumulation of electric current in lead-up to solar flares

    AU - Dhuri, Dattaraj B.

    AU - Hanasoge, Shravan

    AU - Cheung, Mark C.M.

    PY - 2019/6/4

    Y1 - 2019/6/4

    N2 - Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.

    AB - Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.

    KW - Machine learning

    KW - Solar flares

    KW - Solar magnetic fields

    KW - Space weather

    UR - http://www.scopus.com/inward/record.url?scp=85066614522&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85066614522&partnerID=8YFLogxK

    U2 - 10.1073/pnas.1820244116

    DO - 10.1073/pnas.1820244116

    M3 - Article

    VL - 166

    SP - 11141

    EP - 11146

    JO - Proceedings of the National Academy of Sciences of the United States of America

    JF - Proceedings of the National Academy of Sciences of the United States of America

    SN - 0027-8424

    IS - 23

    ER -