Ionospheric Prediction With High Temporal Resolution Using a Local Data Ingestion Technique Over the Chinese Region

The high temporal resolution of ionospheric predictions is essential for monitoring ionospheric activity due to its rapid and complex time-variable effects. The data ingestion technique, as a potential candidate, facilitates comprehensive ionospheric predictions with high temporal resolution for local regions. In this study, total electron content (TEC) observations from local reference stations were input into the NeQuick-2 background model to enhance its performance. The accuracy of ionospheric predictions at a 5-min resolution with the updated NeQuick-2 model was investigated for a selected geomagnetic storm period in Solar Cycle 25. The results indicate that the updated NeQuick-2 model with a 5-min resolution can provide reliable ionospheric predictions, although the model accuracy is somewhat affected during storm time. Specifically, the TEC accuracy of the updated model is comparable to that of the IGS global ionospheric model but with smaller and more stable offsets. Compared to the standard NeQuick-2 model, the updated model demonstrates a significant reduction in both the bias and root-mean-square error (RMSE) from approximately 2 TECu to approximately 0.2 TECu and from approximately 8 TECu to nearly 4 TECu, respectively. For frequency of the F2 layer, the updated model outperforms the standard model with a smaller bias and a decreased RMSE from approximately 1.3 MHz to approximately 1.0 MHz. The electron density profiles from the updated model align more closely with the COSMIC-2 observations than those from the standard model. Furthermore, compared with the hourly updated model, the 5-min updated model slightly improved the accuracy of the TEC predictions.