We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Reconstruction of the Regional Total Electron Content Maps Over the Korean Peninsula Using Deep Convolutional Generative Adversarial Network and Poisson Blending.
- Authors
Jeong, Se‐Heon; Lee, Woo Kyoung; Jang, Soojeong; Kil, Hyosub; Kim, Jeong‐Heon; Kwak, Young‐Sil; Kim, Yong Ha; Hong, Junseok; Choi, Byung‐Kyu
- 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. Plain Language Summary: Total electron content (TEC) is a parameter that represents the column number density of ionospheric plasma. This parameter is a valuable resource for the study of ionospheric phenomena and the prediction of space weather caused by the ionosphere. The dense ground‐based network of Global Navigation Satellite System (GNSS) stations provides the means to generate global TEC maps which are essential for the continuous monitoring of the ionosphere. However, data are always missing, especially in the ocean area, due to the geographical limitations of ground‐based GNSS stations. Recently, several studies applied deep learning techniques to overcome the shortcomings in the construction of global TEC maps. However, deep learning techniques have not yet been tested for the reconstruction of high‐resolution regional TEC maps. By applying the deep learning techniques developed in previous studies, we successfully reconstructed regional TEC maps in the vicinity of the Korean Peninsula. Our results demonstrate that deep learning techniques can contribute to near real‐time monitoring of the ionosphere by immediately providing complete, high‐resolution TEC maps. Key Points: Local total electron content maps are completed with deep convolutional generative adversarial network and Poisson blending (DCGAN‐PB)The DCGAN‐PB model satisfactorily fills data gaps and reconstructs fine ionospheric structures in the vicinity of the Korean PeninsulaThe DCGAN‐PB model performance is not sensitive to observed data number and geophysical parameters
- Subjects
KOREA; GENERATIVE adversarial networks; GLOBAL Positioning System; IONOSPHERIC plasma
- Publication
Space Weather: The International Journal of Research & Applications, 2022, Vol 20, Issue 8, p1
- ISSN
1539-4956
- Publication type
Article
- DOI
10.1029/2022SW003131