Potential of regional ionosphere prediction using a long short-term
memory (LSTM) deep-learning algorithm specialized for geomagnetic storm
period
Abstract
In our previous study (Moon et al., 2020), we developed a Long Short
Term Memory (LSTM) deep-learning model for geomagnetic quiet days to
perform effective long-term predictions for the regional ionosphere.
However, their model could not predict geomagnetic storm days
effectively at all. This study developed an LSTM model suitable for
geomagnetic storms using the new training data set and re-designing
input parameters and hyper-parameters. We collected 131 days of
geomagnetic storm cases from 1 January 2009 to 31 December 2019, and
obtained the IMF Bz, Dst, Kp, and AE indices related to the geomagnetic
storm corresponding to each storm date. These indices and F2 parameters
of Jeju ionosonde (33.43˚N, 126.30˚E) were used as input parameters for
the LSTM model. To test and verify the predictive performance and the
usability of the LSTM model for geomagnetic storms developed in this
manner, we created and diagnosed the 0.5, 1, 2, 3, 6, 12, and 24-hour
predictive LSTM models. According to the results of this study, the LSTM
storm model for 24-hour developed in this study achieved a predictive
performance during the geomagnetic storms about 32% (10%), 34%
(17%), and 37% (5%) better in RMSE of foF2 (hmF2) than the LSTM quiet
model (Moon et al., 2020), SAMI2, and IRI-2016 models. We propose that
the short-term predictions of less than 3 hours are sufficiently
competitive compared to other traditional ionospheric models. Thus, this
study suggests that our model can be used for short-term prediction and
monitoring of the regional mid-latitude ionosphere.