Two Hours Ahead Prediction of the TEC over China Using a Deep Learning Method

In this paper, we forecast the regional total electron content (TEC) over China (0–60° N, 70–140° E) two hours in advance using a deep learning method called pix2pixhd that is based on Generative Adversarial Networks (GAN). We use International GNSS Service (IGS) TEC maps during 2003–2018 period for training and divide data into three parts: set (2003–2013), test (2014–2017), validation (2018). evaluate prediction effect of our model Root Mean Square Error correlation coefficient compare with IRI-2016. The result demonstrates shows good performance China. Under different geomagnetic solar activity conditions, always better than Analyzing average difference map between output target IGS (+2 h), behaves well including low-latitude region. addition, quiet time high years. successful application forecasting methods can solve many geoscience problems, especially ionospheric parameter forecasting.