Search for: All records

Abstract This study evaluates the performance of deep learning approach in the prediction of the ionospheric total electron content (TEC) during magnetically quiet periods. Two deep learning techniques, long short‐term memory (LSTM) and convolutional LSTM (ConvLSTM), are employed to predict TEC values 24 hr ahead in the vicinity of the Korean Peninsula (26.5°–40°N, 121°–134.5°E). The LSTM method predicts TEC at a single point based on time series of data at that point, whereas the ConvLSTM method simultaneously predicts TEC values at multiple points using spatiotemporal distribution of TEC. Both the LSTM and ConvLSTM models are trained using the complete regional TEC maps reconstructed by applying the Deep Convolutional Generative Adversarial Network–Poisson Blending (DCGAN‐PB) method to observed TEC data. The training period spans from 2002 to 2018, and the model performance is evaluated using 2019 data. Our results show that the ConvLSTM method outperforms the LSTM method, generating more reliable TEC maps with smaller root mean square errors when compared to the ground truth (DCGAN‐PB TEC maps). This outcome indicates that deep learning models can improve the prediction accuracy of TEC at a specific point by taking into account spatial information of TEC. We conclude that ConvLSTM is a reliable and efficient approach for the prompt ionospheric prediction. 

Abstract This study reconstructs total electron content (TEC) maps in the vicinity of the Korean Peninsula by employing a deep convolutional generative adversarial network and Poisson blending (DCGAN‐PB). Our interest is to rebuild small‐scale ionosphere structures on the TEC map in a local region where pronounced ionospheric structures, such as the equatorial ionization anomaly, are absent. The reconstructed regional TEC maps have a domain of 120°–135.5°E longitude and 25.5°–41°N latitude with 0.5° resolution. To achieve this, we first train a DCGAN model by using the International Reference Ionosphere‐based TEC maps from 2002 to 2019 (except for 2010 and 2014) as a training data set. Next, the trained DCGAN model generates synthetic complete TEC maps from observation‐based incomplete TEC maps. Final TEC maps are produced by blending of synthetic TEC maps with observed TEC data by PB. The performance of the DCGAN‐PB model is evaluated by testing the regeneration of the masked TEC observations in 2010 (solar minimum) and 2014 (solar maximum). Our results show that a good correlation between the masked and model‐generated TEC values is maintained even with a large percentage (∼80%) of masking. The performance of the DCGAN‐PB model is not sensitive to local time, solar activity, and magnetic activity. Thus, the DCGAN‐PB model can reconstruct fine ionospheric structures in regions where observations are sparse and distinguishing ionospheric structures are absent. This model can contribute to near real‐time monitoring of the ionosphere by immediately providing complete TEC maps.