Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective
Abstract Several techniques have been developed in the last two decades to forecast the occurrence of Solar Proton Events (SPEs), mainly based on the statistical association between the >10 MeV proton flux and precursor parameters. The Empirical model for Solar Proton Events Real Time Alert (ESPE...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-10-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2021SW002794 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536439549952000 |
|---|---|
| author | Mirko Stumpo Simone Benella Monica Laurenza Tommaso Alberti Giuseppe Consolini Maria Federica Marcucci |
| author_facet | Mirko Stumpo Simone Benella Monica Laurenza Tommaso Alberti Giuseppe Consolini Maria Federica Marcucci |
| author_sort | Mirko Stumpo |
| collection | DOAJ |
| description | Abstract Several techniques have been developed in the last two decades to forecast the occurrence of Solar Proton Events (SPEs), mainly based on the statistical association between the >10 MeV proton flux and precursor parameters. The Empirical model for Solar Proton Events Real Time Alert (ESPERTA; Laurenza et al., 2009, https://doi.org/10.1029/2007sw000379) provides a quite good and timely prediction of SPEs after the occurrence of ≥M2 soft x‐ray (SXR) bursts, by using as input parameters the flare heliolongitude, the SXR and the ∼1 MHz radio fluence. Here, we reinterpret the ESPERTA model in the framework of machine learning and perform a cross validation, leading to a comparable performance. Moreover, we find that, by applying a cut‐off on the ≥M2 flares heliolongitude, the False Alarm Rate (FAR) is reduced. The cut‐off is set to E20° where the cumulative distribution of ≥M2 flares associated with SPEs shows a break which reflects the poor magnetic connection between the Earth and eastern hemisphere flares. The best performance is obtained by using the SMOTE algorithm, leading to probability of detection of 0.83 and a FAR of 0.39. Nevertheless, we demonstrate that a relevant FAR on the predictions is a natural consequence of the sample base rates. From a Bayesian point of view, we find that the FAR explicitly contains the prior knowledge about the class distributions. This is a critical issue of any statistical approach, which requires to perform the model validation by preserving the class distributions within the training and test datasets. |
| format | Article |
| id | doaj-art-ce4ad144981e448798e07d39d391a048 |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2021-10-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-ce4ad144981e448798e07d39d391a0482025-01-14T16:30:35ZengWileySpace Weather1542-73902021-10-011910n/an/a10.1029/2021SW002794Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning PerspectiveMirko Stumpo0Simone Benella1Monica Laurenza2Tommaso Alberti3Giuseppe Consolini4Maria Federica Marcucci5INAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyAbstract Several techniques have been developed in the last two decades to forecast the occurrence of Solar Proton Events (SPEs), mainly based on the statistical association between the >10 MeV proton flux and precursor parameters. The Empirical model for Solar Proton Events Real Time Alert (ESPERTA; Laurenza et al., 2009, https://doi.org/10.1029/2007sw000379) provides a quite good and timely prediction of SPEs after the occurrence of ≥M2 soft x‐ray (SXR) bursts, by using as input parameters the flare heliolongitude, the SXR and the ∼1 MHz radio fluence. Here, we reinterpret the ESPERTA model in the framework of machine learning and perform a cross validation, leading to a comparable performance. Moreover, we find that, by applying a cut‐off on the ≥M2 flares heliolongitude, the False Alarm Rate (FAR) is reduced. The cut‐off is set to E20° where the cumulative distribution of ≥M2 flares associated with SPEs shows a break which reflects the poor magnetic connection between the Earth and eastern hemisphere flares. The best performance is obtained by using the SMOTE algorithm, leading to probability of detection of 0.83 and a FAR of 0.39. Nevertheless, we demonstrate that a relevant FAR on the predictions is a natural consequence of the sample base rates. From a Bayesian point of view, we find that the FAR explicitly contains the prior knowledge about the class distributions. This is a critical issue of any statistical approach, which requires to perform the model validation by preserving the class distributions within the training and test datasets.https://doi.org/10.1029/2021SW002794Machine learningsolar flaressolar proton eventsspace weatherstatistical SPE forecasting |
| spellingShingle | Mirko Stumpo Simone Benella Monica Laurenza Tommaso Alberti Giuseppe Consolini Maria Federica Marcucci Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective Space Weather Machine learning solar flares solar proton events space weather statistical SPE forecasting |
| title | Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective |
| title_full | Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective |
| title_fullStr | Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective |
| title_full_unstemmed | Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective |
| title_short | Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective |
| title_sort | open issues in statistical forecasting of solar proton events a machine learning perspective |
| topic | Machine learning solar flares solar proton events space weather statistical SPE forecasting |
| url | https://doi.org/10.1029/2021SW002794 |
| work_keys_str_mv | AT mirkostumpo openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective AT simonebenella openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective AT monicalaurenza openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective AT tommasoalberti openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective AT giuseppeconsolini openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective AT mariafedericamarcucci openissuesinstatisticalforecastingofsolarprotoneventsamachinelearningperspective |