Author-submitted data information
| ID | 620 |
| Title | Data of Spatio-Temporal deep learning model for regional EPB irregularities short-term Prediction |
| Creator | Xiukuan Zhao |
| Subject | Equatorial plasma bubble, Ionospheric irregularity, Spatio-temporal deep learning, Prediction model |
| Publisher | Xiukuan Zhao |
| Description | Using the dense ground-based GNSS receiver network and ionosonde data from East and Southeast Asia during 2010-2021, a novel Spatio-Temporal deep learning model for regional EPB irregularities short-term Prediction (STEP) was developed. The model integrates the convolutional neural network (CNN) and long short-term memory (LSTM) network, together with attention mechanisms, to capture both spatial and temporal features of regional ionospheric irregularities. This dataset includes both the model and the results generated by STEP. The parameters provided are: UT (hours), Latitude (°), Longitude (°), Date, Y_pred (TECU/min), and Y_true (TECU/min). The dimensions of Y_pred and Y_true are 10812 x 610, where 10812 represents the product of the number of date and the number of UT (minus 18), and 610 corresponds to the product of the number of Latitude and Longitude. The model with a .pth extension can be loaded using PyTorch. |
| Contributor | Guozhu Li, Haiyong Xie, Lianhuan Hu, Wenjie Sun, Yi Li, Guofeng Dai, Jianfei Liu, Yu Li, Baiqi Ning, Michi Nishioka, Septi Perwitasari, and Prasert Kenpankho |
| Date | 2010-2021 |
| Type | The parameters include UT (hour), Latitude (°), Longitude (°), Date, Y_pred (TECU/min) and Y_true (TECU/min). |
| Format | the .pth extension can be loaded using PyTorch, while .mat files are compatible with MATLAB |
| URL | http://www.geophys.ac.cn/ArticleData/20241018STEPdata.zip |
| DOI | 10.12197/2024GA021 |
| Source | |
| Language | eng |
| Relation | |
| Coverage | |
| Rights | Institute of Geology and Geophysics, Chinese Academy of Sciences |