Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds
Abstract Large geomagnetically induced currents (GICs) pose a risk to ground based infrastructure such as power networks. Large GICs may be induced when the rate of change of the ground magnetic field is significantly elevated. We assess the ability of three different machine learning model architec...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-09-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2021SW002788 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536482324512768 |
|---|---|
| author | A. W. Smith C. Forsyth I. J. Rae T. M. Garton T. Bloch C. M. Jackman M. Bakrania |
| author_facet | A. W. Smith C. Forsyth I. J. Rae T. M. Garton T. Bloch C. M. Jackman M. Bakrania |
| author_sort | A. W. Smith |
| collection | DOAJ |
| description | Abstract Large geomagnetically induced currents (GICs) pose a risk to ground based infrastructure such as power networks. Large GICs may be induced when the rate of change of the ground magnetic field is significantly elevated. We assess the ability of three different machine learning model architectures to process the time history of the incoming solar wind and provide a probabilistic forecast as to whether the rate of change of the ground magnetic field will exceed specific high thresholds at a location in the UK. The three models tested represent feed forward, convolutional and recurrent neural networks. We find all three models are reliable and skillful, with Brier skill scores, receiver‐operating characteristic scores and precision‐recall scores of approximately 0.25, 0.95 and 0.45, respectively. When evaluated during two example magnetospheric storms we find that all scores increase significantly, indicating that the models work better during active intervals. The models perform excellently through the majority of the storms, however they do not fully capture the ground response around the initial sudden commencements. We attribute this to the use of propagated solar wind data not allowing the models notice to forecast impulsive phenomenon. Increasing the volume of solar wind data provided to the models does not produce appreciable increases in model performance, possibly due to the fixed model structures and limited training data. However, increasing the horizon of the forecast from 30 min to 3 h increases the performance of the models, presumably as the models need not be as precise about timing. |
| format | Article |
| id | doaj-art-56247dda579e423da6002dbccd56ee27 |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2021-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-56247dda579e423da6002dbccd56ee272025-01-14T16:26:53ZengWileySpace Weather1542-73902021-09-01199n/an/a10.1029/2021SW002788Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and ThresholdsA. W. Smith0C. Forsyth1I. J. Rae2T. M. Garton3T. Bloch4C. M. Jackman5M. Bakrania6Mullard Space Science Laboratory UCL Dorking UKMullard Space Science Laboratory UCL Dorking UKDepartment of Mathematics, Physics and Electrical Engineering Northumbria University Newcastle upon Tyne UKSpace Environment Physics Group Department of Physics and Astronomy University of Southampton Southampton UKDepartment of Meteorology University of Reading Reading UKSchool of Cosmic Physics DIAS Dunsink Observatory Dublin Institute for Advanced Studies Dublin IrelandMullard Space Science Laboratory UCL Dorking UKAbstract Large geomagnetically induced currents (GICs) pose a risk to ground based infrastructure such as power networks. Large GICs may be induced when the rate of change of the ground magnetic field is significantly elevated. We assess the ability of three different machine learning model architectures to process the time history of the incoming solar wind and provide a probabilistic forecast as to whether the rate of change of the ground magnetic field will exceed specific high thresholds at a location in the UK. The three models tested represent feed forward, convolutional and recurrent neural networks. We find all three models are reliable and skillful, with Brier skill scores, receiver‐operating characteristic scores and precision‐recall scores of approximately 0.25, 0.95 and 0.45, respectively. When evaluated during two example magnetospheric storms we find that all scores increase significantly, indicating that the models work better during active intervals. The models perform excellently through the majority of the storms, however they do not fully capture the ground response around the initial sudden commencements. We attribute this to the use of propagated solar wind data not allowing the models notice to forecast impulsive phenomenon. Increasing the volume of solar wind data provided to the models does not produce appreciable increases in model performance, possibly due to the fixed model structures and limited training data. However, increasing the horizon of the forecast from 30 min to 3 h increases the performance of the models, presumably as the models need not be as precise about timing.https://doi.org/10.1029/2021SW002788GICsspace weatherforecastingmachine learningneural networksmodel validation |
| spellingShingle | A. W. Smith C. Forsyth I. J. Rae T. M. Garton T. Bloch C. M. Jackman M. Bakrania Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds Space Weather GICs space weather forecasting machine learning neural networks model validation |
| title | Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds |
| title_full | Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds |
| title_fullStr | Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds |
| title_full_unstemmed | Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds |
| title_short | Forecasting the Probability of Large Rates of Change of the Geomagnetic Field in the UK: Timescales, Horizons, and Thresholds |
| title_sort | forecasting the probability of large rates of change of the geomagnetic field in the uk timescales horizons and thresholds |
| topic | GICs space weather forecasting machine learning neural networks model validation |
| url | https://doi.org/10.1029/2021SW002788 |
| work_keys_str_mv | AT awsmith forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT cforsyth forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT ijrae forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT tmgarton forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT tbloch forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT cmjackman forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds AT mbakrania forecastingtheprobabilityoflargeratesofchangeofthegeomagneticfieldintheuktimescaleshorizonsandthresholds |