One‐Day Forecasting of Global TEC Using a Novel Deep Learning Model

Lee, Sujin; Ji, Eun‐Young; Moon, Yong‐Jae; Park, Eunsu

doi:10.1029/2020sw002600

Cited by 40 publications

(30 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wang et al (2020) directly used an adaptive autoregressive model to predict grid point vertical total electron content values, avoiding twice accumulated errors by predicting spherical harmonic coefficients. Lee et al (2020) used a conditional generative adversarial network to predict the global TEC maps one day in advance, the RMSE from the model is 1.74 TECU.…”

mentioning

confidence: 99%

ED‐ConvLSTM: A Novel Global Ionospheric Total Electron Content Medium‐Term Forecast Model

et al. 2022

View full text Add to dashboard Cite

In this paper, we proposed an innovative encoder‐decoder structure with a convolution long short‐term memory (ED‐ConvLSTM) network to forecast global total electron content (TEC) based on the International GNSS Service (IGS) TEC maps from 2005 to 2018 with 1‐hr time cadence. The ED‐ConvLSTM model is used to forecast TEC maps 1–7 days in advance through iterations. To investigate the model's performance, we compared the model with International Reference Ionosphere (IRI2016) model in 2014 and 2018, and compared the model with 1‐day Beijing University of Aeronautics and Astronautics (BUAA) model in 2018. The results show that our 7‐day ED‐ConvLSTM model (ED‐ConvLSTM model that forecasts 7 days in advance) outperforms IRI2016 in 2014 and 2018, and our 5‐day ED‐ConvLSTM model (ED‐ConvLSTM model that forecasts 5 days in advance) outperforms 1‐day BUAA model. Furthermore, the root mean square error (RMSE) from the 1‐day ED‐ConvLSTM model with respect to the IGS TEC maps decreases by 51.5% and 43%, respectively, in 2014 and 2018 compared with that from IRI2016 model. The RMSE from the 1‐day ED‐ConvLSTM model is 20.3% lower than that from the 1‐day BUAA model in 2018. In addition, our model has the highest RMSE in the Equatorial Ionospheric Anomaly (EIA) region, but can roughly predict the features and locations of EIA. However, the model fails to forecast localized TEC enhancement and the sudden ionospheric response to the geomagnetic storms. Overall, the model shows competitive performance in medium‐term global TEC maps prediction during geomagnetic quiet periods.

show abstract

mentioning

confidence: 99%

ED‐ConvLSTM: A Novel Global Ionospheric Total Electron Content Medium‐Term Forecast Model

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Several existing maps were used as references to interpolate missing values in some regions, such as the oceans. The TEC maps can also be predicted two hours in advance with an LSTM (Liu et al, 2020) or one day in advance with a GAN (Lee et al, 2021). Further, a DNN is used to estimate the relationship between electron temperature and electron density in small regions (Hu et al, 2020 Note.…”

Section: Space Sciencementioning

confidence: 99%

Deep Learning for Geophysics: Current and Future Trends

2021

Reviews of Geophysics

281

View full text Add to dashboard Cite

show abstract

“…The DL technique based on conditional generative adversarial networks (cGAN) is used to forecast global TEC maps one-day ahead [8]. Saed et al have proposed a solar flare predictions model using the ML technique, the Support Vector Machine (SVM) model with Center for Orbit Determination in Europe (CODE) Global Ionospheric TEC maps (GIM) TEC data.…”

Section: Introductionmentioning

confidence: 99%

A Bidirectional Deep-Learning Algorithm to Forecast Regional Ionospheric TEC Maps

Krishna

Ratnam

Sivavaraprasad

2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

The rapid evolutions in Artificial Intelligence (AI) and the Machine Learning era have significantly improved accuracy for ionospheric space weather forecasting models. The ionospheric Total Electron Content (TEC) forecasting is necessary to alert Global Navigation Satellite System (GNSS) users about ionospheric space weather influences on satellitereceiver radio communications. Precise modeling and forecasting of the ionospheric TEC are critical for reliable and accurate GNSS applications. In this paper, a deep learning-based Bidirectional Long Short-Term Memory (Bi-LSTM) algorithm is implemented for 26 GPS (Global Positioning System) stations TEC data over the Indian region. The Bi-LSTM ionospheric TEC forecasting maps are generated and compared with ANN, and LSTM models during both geomagnetic quiet and disturbed periods. The potential of Bi-LSTM networks in time sequence processing is improved by having forward and backward connections. The Bi-LSTM results are demonstrated with and without solar and geomagnetic indices as input to the network. This work's outcome would help develop an ionospheric weather alert system for GNSS users.

show abstract

One‐Day Forecasting of Global TEC Using a Novel Deep Learning Model

Cited by 40 publications

References 20 publications

ED‐ConvLSTM: A Novel Global Ionospheric Total Electron Content Medium‐Term Forecast Model

ED‐ConvLSTM: A Novel Global Ionospheric Total Electron Content Medium‐Term Forecast Model

Deep Learning for Geophysics: Current and Future Trends

A Bidirectional Deep-Learning Algorithm to Forecast Regional Ionospheric TEC Maps

Contact Info

Product

Resources

About