Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method
Abstract The low Earth orbit (LEO) satellite provides valuable direct observations for scientific investigation of the plasmasphere, while the plasmaspheric total electron content (PTEC) with a high temporal resolution cannot be precisely estimated due to fewer LEO satellites. In this paper, a novel...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-07-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2022SW003347 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536444202483712 |
|---|---|
| author | Qiqi Shi Shuanggen Jin Linlin Li |
| author_facet | Qiqi Shi Shuanggen Jin Linlin Li |
| author_sort | Qiqi Shi |
| collection | DOAJ |
| description | Abstract The low Earth orbit (LEO) satellite provides valuable direct observations for scientific investigation of the plasmasphere, while the plasmaspheric total electron content (PTEC) with a high temporal resolution cannot be precisely estimated due to fewer LEO satellites. In this paper, a novel joint method of radial basis function neural network—Kriging (RBF‐Kr) method is designed to construct the daily PTEC model using the Constellation Observing System for Meteorology, Ionosphere and Climate Global Positioning System observations during the low (2009) and high (2013) solar activity years. Compared with the original RBF method, the RBF‐Kr method reduces the mean absolute error and root mean square error from 0.77 to 0.60 TEC unit (TECU) and 0.99 to 0.80 TECU, respectively. The correlation coefficient (Corr) increased from 0.90 to 0.94. Furthermore, daily PTEC variations show that the PTEC at low latitudes is evenly distributed during equinox periods. The South American‐Atlantic Ocean sector has a peak and trough in PTEC during the December and June solstices. A certain symmetrical distribution of PTEC is observed in the latitudinal direction, and the symcenter moves toward the summer hemisphere. The duration of extremal PTEC at 60°W is observed, which lasted up to more than 80 days around the December solstice. An obvious correlation between the solar flux and PTEC is found with up to 0.86, indicating that daily PTEC variations are mainly related to solar activities. |
| format | Article |
| id | doaj-art-dd67d85bd27a47948045a56145d4ea2e |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2023-07-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-dd67d85bd27a47948045a56145d4ea2e2025-01-14T16:27:34ZengWileySpace Weather1542-73902023-07-01217n/an/a10.1029/2022SW003347Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging MethodQiqi Shi0Shuanggen Jin1Linlin Li2Shanghai Astronomical Observatory Chinese Academy of Sciences Shanghai ChinaShanghai Astronomical Observatory Chinese Academy of Sciences Shanghai ChinaShanghai Astronomical Observatory Chinese Academy of Sciences Shanghai ChinaAbstract The low Earth orbit (LEO) satellite provides valuable direct observations for scientific investigation of the plasmasphere, while the plasmaspheric total electron content (PTEC) with a high temporal resolution cannot be precisely estimated due to fewer LEO satellites. In this paper, a novel joint method of radial basis function neural network—Kriging (RBF‐Kr) method is designed to construct the daily PTEC model using the Constellation Observing System for Meteorology, Ionosphere and Climate Global Positioning System observations during the low (2009) and high (2013) solar activity years. Compared with the original RBF method, the RBF‐Kr method reduces the mean absolute error and root mean square error from 0.77 to 0.60 TEC unit (TECU) and 0.99 to 0.80 TECU, respectively. The correlation coefficient (Corr) increased from 0.90 to 0.94. Furthermore, daily PTEC variations show that the PTEC at low latitudes is evenly distributed during equinox periods. The South American‐Atlantic Ocean sector has a peak and trough in PTEC during the December and June solstices. A certain symmetrical distribution of PTEC is observed in the latitudinal direction, and the symcenter moves toward the summer hemisphere. The duration of extremal PTEC at 60°W is observed, which lasted up to more than 80 days around the December solstice. An obvious correlation between the solar flux and PTEC is found with up to 0.86, indicating that daily PTEC variations are mainly related to solar activities.https://doi.org/10.1029/2022SW003347plasmaspheric TECCOSMICRBF neural networkkriging |
| spellingShingle | Qiqi Shi Shuanggen Jin Linlin Li Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method Space Weather plasmaspheric TEC COSMIC RBF neural network kriging |
| title | Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method |
| title_full | Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method |
| title_fullStr | Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method |
| title_full_unstemmed | Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method |
| title_short | Daily Plasmaspheric TEC Variations From COSMIC GPS Observations Based on RBF Neural Network‐Kriging Method |
| title_sort | daily plasmaspheric tec variations from cosmic gps observations based on rbf neural network kriging method |
| topic | plasmaspheric TEC COSMIC RBF neural network kriging |
| url | https://doi.org/10.1029/2022SW003347 |
| work_keys_str_mv | AT qiqishi dailyplasmaspherictecvariationsfromcosmicgpsobservationsbasedonrbfneuralnetworkkrigingmethod AT shuanggenjin dailyplasmaspherictecvariationsfromcosmicgpsobservationsbasedonrbfneuralnetworkkrigingmethod AT linlinli dailyplasmaspherictecvariationsfromcosmicgpsobservationsbasedonrbfneuralnetworkkrigingmethod |