EGU21-9995
https://doi.org/10.5194/egusphere-egu21-9995
EGU General Assembly 2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Deep Neural Networks With Convolutional and LSTM Layers for SYM-H and ASY-H Forecasting

Armando Collado1, Pablo Muñoz1, and Consuelo Cid2
Armando Collado et al.
  • 1Department of Computer Engineering, University of Alcala, Alcalá de Henares, Spain (armando.collado@uah.es, pablo.munoz@uah.es)
  • 2Department of Physics and Mathematics, University of Alcala, Alcalá de Henares, Spain (consuelo.cid@uah.es)

Geomagnetic indices quantify the disturbance caused by the solar activity in particular regions of the Earth. Among them, the SYM-H and ASY-H indices represent the (longitudinally) symmetric and asymmetric geomagnetic disturbance of the horizontal component of the magnetic field at mid-latitude with a 1-minute resolution. Their resolution, along with their relation to the solar wind parameters, makes the forecasting of the geomagnetic indices a problem that can be addressed through the use of Deep Learning, particularly using Deep Neural Networks (DNN). In this work, we present two DNNs developed to forecast the SYM-H and ASY-H indices. Both networks have been trained using solar wind data from the last two solar cycles and they are able to accurately forecast the indices two hours in advance, considering the solar wind and indices values for the previous 16 hours. The evaluation of both networks reveals a great precision for the forecasting, including good predictions for large storms that occurred during the solar cycle 23.

How to cite: Collado, A., Muñoz, P., and Cid, C.: Deep Neural Networks With Convolutional and LSTM Layers for SYM-H and ASY-H Forecasting, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-9995, https://doi.org/10.5194/egusphere-egu21-9995, 2021.

Displays

Display file