1. The Response of Ionospheric Currents to External Drivers Investigated Using a Neural Network‐Based Model.
- Author
-
Cao, Xin, Chu, Xiangning, Bortnik, Jacob, Weygand, James M., Li, Jinxing, Aryan, Homayon, and Ma, Donglai
- Subjects
FEEDFORWARD neural networks ,SOLAR cycle ,SOLAR radiation ,MODERN society ,ELECTRIC power distribution grids ,PREDICTION models - Abstract
A predictive model for the variation of ionospheric currents is of great scientific and practical importance to our modern industrial society. To study the response of ionospheric currents to external drivers including geomagnetic indices and solar radiation, we developed a feedforward neural network model trained on the Equivalent Ionospheric Current (EIC) data from 1st January 2007 to 31st December 2019. Due to the highly imbalanced nature of the ionospheric currents data, which means that the data of extreme events are much less than those of quiet times, we utilized different loss functions to improve the model performance. Our model demonstrates the potential to predict the active events of ionospheric currents reasonably well (e.g., EICs during substorms) within a timescale of a few minutes. Although the data used for training are measurements over the North American and Greenland sectors, our model is not only able to predict EICs within this region, but is also able to provide a promising out‐of‐sample prediction on a global scale. Plain Language Summary: Studying the ionospheric current system is important for our modern society, since its drastic change can impact the ground‐based facilities such as power grid. We used a neural network model to predict the influence of geomagnetic activities and upstream condition on the variation of ionospheric currents over the North American and Greenland regions during a whole solar cycle period. Because the data set of ionospheric currents is highly imbalanced, we used different loss function to optimize our model. We found that our model can reasonably well predict the ionospheric current events which are related to the magnetospheric activities. We also found that our model can provide predictions beyond the North American and Greenland regions. Key Points: A neural network is trained to predict the response of Equivalent Ionospheric Currents (EIC) to geomagnetic indicesA comparative study between the mean square error and focal loss has been made to address the EIC's imbalance regression problemOur model can spatially and temporally predict the eastward‐westward EIC component [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF