Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria
Abstract The forecasting of local GIC effects has largely relied on the forecasting of dB/dt as a proxy and, to date, little attention has been paid to directly forecasting the geoelectric field or GICs themselves. We approach this problem with machine learning tools, specifically recurrent neural n...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-03-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2021SW002907 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536357352079360 |
|---|---|
| author | R. L. Bailey R. Leonhardt C. Möstl C. Beggan M. A. Reiss A. Bhaskar A. J. Weiss |
| author_facet | R. L. Bailey R. Leonhardt C. Möstl C. Beggan M. A. Reiss A. Bhaskar A. J. Weiss |
| author_sort | R. L. Bailey |
| collection | DOAJ |
| description | Abstract The forecasting of local GIC effects has largely relied on the forecasting of dB/dt as a proxy and, to date, little attention has been paid to directly forecasting the geoelectric field or GICs themselves. We approach this problem with machine learning tools, specifically recurrent neural networks or LSTMs by taking solar wind observations as input and training the models to predict two different kinds of output: first, the geoelectric field components Ex and Ey; and second, the GICs in specific substations in Austria. The training is carried out on the geoelectric field and GICs modeled from 26 years of one‐minute geomagnetic field measurements, and results are compared to GIC measurements from recent years. The GICs are generally predicted better by an LSTM trained on values from a specific substation, but only a fraction of the largest GICs are correctly predicted. This model has a correlation with measurements of around 0.6, and a root‐mean‐square error of 0.7 A. The probability of detecting mild activity in GICs is around 50%, and 15% for larger GICs. |
| format | Article |
| id | doaj-art-bf7c44ea7325464eb7a82916ead7593a |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2022-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-bf7c44ea7325464eb7a82916ead7593a2025-01-14T16:30:58ZengWileySpace Weather1542-73902022-03-01203n/an/a10.1029/2021SW002907Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in AustriaR. L. Bailey0R. Leonhardt1C. Möstl2C. Beggan3M. A. Reiss4A. Bhaskar5A. J. Weiss6Conrad Observatory Zentralanstalt für Meteorologie und Geodynamik Vienna AustriaConrad Observatory Zentralanstalt für Meteorologie und Geodynamik Vienna AustriaSpace Research Institute Austrian Academy of Sciences Graz AustriaBritish Geological Survey Edinburgh UKSpace Research Institute Austrian Academy of Sciences Graz AustriaSpace Physics Laboratory ISRO/Vikram Sarabhai Space Centre Trivandrum IndiaSpace Research Institute Austrian Academy of Sciences Graz AustriaAbstract The forecasting of local GIC effects has largely relied on the forecasting of dB/dt as a proxy and, to date, little attention has been paid to directly forecasting the geoelectric field or GICs themselves. We approach this problem with machine learning tools, specifically recurrent neural networks or LSTMs by taking solar wind observations as input and training the models to predict two different kinds of output: first, the geoelectric field components Ex and Ey; and second, the GICs in specific substations in Austria. The training is carried out on the geoelectric field and GICs modeled from 26 years of one‐minute geomagnetic field measurements, and results are compared to GIC measurements from recent years. The GICs are generally predicted better by an LSTM trained on values from a specific substation, but only a fraction of the largest GICs are correctly predicted. This model has a correlation with measurements of around 0.6, and a root‐mean‐square error of 0.7 A. The probability of detecting mild activity in GICs is around 50%, and 15% for larger GICs.https://doi.org/10.1029/2021SW002907geomagnetically induced currentsneural networksLSTMspace weatherforecasting |
| spellingShingle | R. L. Bailey R. Leonhardt C. Möstl C. Beggan M. A. Reiss A. Bhaskar A. J. Weiss Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria Space Weather geomagnetically induced currents neural networks LSTM space weather forecasting |
| title | Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria |
| title_full | Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria |
| title_fullStr | Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria |
| title_full_unstemmed | Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria |
| title_short | Forecasting GICs and Geoelectric Fields From Solar Wind Data Using LSTMs: Application in Austria |
| title_sort | forecasting gics and geoelectric fields from solar wind data using lstms application in austria |
| topic | geomagnetically induced currents neural networks LSTM space weather forecasting |
| url | https://doi.org/10.1029/2021SW002907 |
| work_keys_str_mv | AT rlbailey forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT rleonhardt forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT cmostl forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT cbeggan forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT mareiss forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT abhaskar forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria AT ajweiss forecastinggicsandgeoelectricfieldsfromsolarwinddatausinglstmsapplicationinaustria |