Incorporating polar field data for improved solar flare prediction
In this paper, we consider incorporating data associated with the sun’s north and south polar field strengths to improve solar flare prediction performance using machine learning models. When used to supplement local data from active regions on the photospheric magnetic field of the sun, the polar f...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2022-12-01
|
Series: | Frontiers in Astronomy and Space Sciences |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fspas.2022.1040107/full |
_version_ | 1797979808562413568 |
---|---|
author | Mehmet Aktukmak Zeyu Sun Monica Bobra Tamas Gombosi Ward B. Manchester IV Yang Chen Alfred Hero Alfred Hero |
author_facet | Mehmet Aktukmak Zeyu Sun Monica Bobra Tamas Gombosi Ward B. Manchester IV Yang Chen Alfred Hero Alfred Hero |
author_sort | Mehmet Aktukmak |
collection | DOAJ |
description | In this paper, we consider incorporating data associated with the sun’s north and south polar field strengths to improve solar flare prediction performance using machine learning models. When used to supplement local data from active regions on the photospheric magnetic field of the sun, the polar field data provides global information to the predictor. While such global features have been previously proposed for predicting the next solar cycle’s intensity, in this paper we propose using them to help classify individual solar flares. We conduct experiments using HMI data employing four different machine learning algorithms that can exploit polar field information. Additionally, we propose a novel probabilistic mixture of experts model that can simply and effectively incorporate polar field data and provide on-par prediction performance with state-of-the-art solar flare prediction algorithms such as the Recurrent Neural Network (RNN). Our experimental results indicate the usefulness of the polar field data for solar flare prediction, which can improve Heidke Skill Score (HSS2) by as much as 10.1%1. |
first_indexed | 2024-04-11T05:45:44Z |
format | Article |
id | doaj.art-769d22232ee146debf842eb010a8eb86 |
institution | Directory Open Access Journal |
issn | 2296-987X |
language | English |
last_indexed | 2024-04-11T05:45:44Z |
publishDate | 2022-12-01 |
publisher | Frontiers Media S.A. |
record_format | Article |
series | Frontiers in Astronomy and Space Sciences |
spelling | doaj.art-769d22232ee146debf842eb010a8eb862022-12-22T04:42:15ZengFrontiers Media S.A.Frontiers in Astronomy and Space Sciences2296-987X2022-12-01910.3389/fspas.2022.10401071040107Incorporating polar field data for improved solar flare predictionMehmet Aktukmak0Zeyu Sun1Monica Bobra2Tamas Gombosi3Ward B. Manchester IV4Yang Chen5Alfred Hero6Alfred Hero7Department of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, United StatesDepartment of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, United StatesW. W. Hansen Experimental Physics Laboratory, Stanford University, Stanford, CA, United StatesDepartment of Aerospace Engineering, University of Michigan, Ann Arbor, MI, United StatesDepartment of Climate and Space Sciences and Engineering, University of Michigan, Ann Arbor, MI, United StatesDepartment of Statistics, University of Michigan, Ann Arbor, MI, United StatesDepartment of Electrical and Computer Engineering, University of Michigan, Ann Arbor, MI, United StatesDepartment of Statistics, University of Michigan, Ann Arbor, MI, United StatesIn this paper, we consider incorporating data associated with the sun’s north and south polar field strengths to improve solar flare prediction performance using machine learning models. When used to supplement local data from active regions on the photospheric magnetic field of the sun, the polar field data provides global information to the predictor. While such global features have been previously proposed for predicting the next solar cycle’s intensity, in this paper we propose using them to help classify individual solar flares. We conduct experiments using HMI data employing four different machine learning algorithms that can exploit polar field information. Additionally, we propose a novel probabilistic mixture of experts model that can simply and effectively incorporate polar field data and provide on-par prediction performance with state-of-the-art solar flare prediction algorithms such as the Recurrent Neural Network (RNN). Our experimental results indicate the usefulness of the polar field data for solar flare prediction, which can improve Heidke Skill Score (HSS2) by as much as 10.1%1.https://www.frontiersin.org/articles/10.3389/fspas.2022.1040107/fullsolar flare predictionpolar fieldssolar cyclemixture modelingactive regions |
spellingShingle | Mehmet Aktukmak Zeyu Sun Monica Bobra Tamas Gombosi Ward B. Manchester IV Yang Chen Alfred Hero Alfred Hero Incorporating polar field data for improved solar flare prediction Frontiers in Astronomy and Space Sciences solar flare prediction polar fields solar cycle mixture modeling active regions |
title | Incorporating polar field data for improved solar flare prediction |
title_full | Incorporating polar field data for improved solar flare prediction |
title_fullStr | Incorporating polar field data for improved solar flare prediction |
title_full_unstemmed | Incorporating polar field data for improved solar flare prediction |
title_short | Incorporating polar field data for improved solar flare prediction |
title_sort | incorporating polar field data for improved solar flare prediction |
topic | solar flare prediction polar fields solar cycle mixture modeling active regions |
url | https://www.frontiersin.org/articles/10.3389/fspas.2022.1040107/full |
work_keys_str_mv | AT mehmetaktukmak incorporatingpolarfielddataforimprovedsolarflareprediction AT zeyusun incorporatingpolarfielddataforimprovedsolarflareprediction AT monicabobra incorporatingpolarfielddataforimprovedsolarflareprediction AT tamasgombosi incorporatingpolarfielddataforimprovedsolarflareprediction AT wardbmanchesteriv incorporatingpolarfielddataforimprovedsolarflareprediction AT yangchen incorporatingpolarfielddataforimprovedsolarflareprediction AT alfredhero incorporatingpolarfielddataforimprovedsolarflareprediction AT alfredhero incorporatingpolarfielddataforimprovedsolarflareprediction |