Global TEC map forecasting model using deep learning based on conditional generative networks

Abstract:

In this study, we develop a global Total Electron Content (TEC) forecasting model using deep learning based on conditional generative networks(cGAN). For training, we use the IGS final TEC maps, which are high accuracy and reliable quality. Our model is trained with the data from 2003 to 2012 and has two input images (IGS TEC map and difference map between the present day and previous day) and one output image (one-day future map). The model is tested with two data sets: solar maximum (2013~2014) and solar minimum (2017~2018). Then we compare the results of our model with those of 1-day CODE prediction model. Our main results from this study are as follows. First, we successfully apply our model to the forecast of global TEC maps. Second, our model well predicts daily TEC maps using only previous TEC maps. The averaged RMSE, BIAS, and STD between our model TEC maps and IGS TEC ones are 2.74 TECU, -0.32 TECU, and 2.59 TECU, respectively. Third, our model generates some peak structures around equatorial regions. Fourth, our model shows better performance than 1-day CODE prediction model during all test periods. Our results have demonstrated that our deep learning model based on an image translation method will be effective for forecasting of future images using previous data.