A comparison of neural network-based predictions of foF2 with the IRI-2012 model at conjugate points in Southeast Asia

Abstract This paper presents the development of Neural Network (NN) model for the prediction of the F2 layer critical frequency (foF2) at three ionosonde stations near the magnetic equator of Southeast Asia. Two of these stations including Chiang Mai (18.76°N, 98.93°E, dip angle 12.7°N) and Kototabang (0.2°S, 100.3°E, dip angle 10.1°S) are at the conjugate points while Chumphon (10.72°N, 99.37°E, dip angle 3.0°N) station is near the equator. To produce the model, the feed forward network with backpropagation algorithm is applied. The NN is trained with the daily hourly values of foF2 during 2004–2012, except 2009, and the selected input parameters, which affect the foF2 variability, include day number (DN), hour number (HR), solar zenith angle (C), geographic latitude ( θ ), magnetic inclination (I), magnetic declination (D) and angle of meridian (M) relative to the sub-solar point, the 7-day mean of F10.7 (F10.7_7), the 81-day mean of SSN (SSN_81) and the 2-day mean of Ap (Ap_2). The foF2 data of 2009 and 2013 are then used for testing the NN model during the foF2 interpolation and extrapolation, respectively. To examine the performance of the proposed NN, the root mean square error (RMSE) of the observed foF2, the proposed NN model and the IRI-2012 (CCIR and URSI options) model are compared. In general, the results show the same trends in foF2 variation between the models (NN and IRI-2012) and the observations in that they are higher during the day and lower at night. Besides, the results demonstrate that the proposed NN model can predict the foF2 values more closely during daytime than during nighttime as supported by the lower RMSE values during daytime (0.5 ≤ RMSE ≤ 1.0 for Chumphon and Kototabang, 0.7 ≤ RMSE ≤ 1.2 at Chiang Mai) and with the highest levels during nighttime (0.8 ≤ RMSE ≤ 1.5 for Chumphon and Kototabang, 1.2 ≤ RMSE ≤ 2.0 at Chiang Mai). Furthermore, the NN model predicts the foF2 values more accurately than the IRI model at the three sites on average, as clearly seen on the yearly RMSE averages. The RMSE values of NN model are lower than those of both CCIR and URSI options, and in terms of the yearly percentage improvements, the NN model gives improvement of around 10–15% in 2009 and 10% in 2013 for Chiang Mai, 20–25% in 2009 and 5–10% in 2013 for Chumphon, and around 18–25% in 2009 and 20–30% in 2013 at Kototabang. Although the NN model predicts the foF2 values closely to the observed data and produces more accurate prediction than the IRI models, in some cases, the IRI model performs better than the NN model. Hence, there is still room for further improvement of the proposed NN model.

[1]  I. Batista,et al.  Ionospheric variability at Brazilian low and equatorial latitudes: comparison between observations and IRI model , 2004 .

[2]  Pornchai Supnithi,et al.  TEC prediction with neural network for equatorial latitude station in Thailand , 2012, Earth, Planets and Space.

[3]  N. Wichaipanich,et al.  Prediction of foF2 using Neural Network at Thailand equatorial latitude station, Chumphon , 2014, 2014 11th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON).

[4]  Adewale Ao.  Comparison of neural network technique with IRI 2001 model ionospheric predictions during great geomagnetic storms for a mid-latitude station. , 2010 .

[5]  Bodo W. Reinisch,et al.  International Reference Ionosphere 2000 , 2001 .

[6]  O. S. Poltavsky,et al.  Evaluation of the IRI model for the European region , 2009 .

[7]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[8]  E. O. Oyeyemi,et al.  Progress towards a new global foF2 model for the International Reference Ionosphere (IRI) , 2009 .

[9]  T. Xenos.  Neural-network-based prediction techniques for single station modeling and regional mapping of the foF2 and M(3000)F2 ionospheric characteristics , 2002 .

[10]  Haris Haralambous,et al.  A Neural Network for Spatial and Temporal Modeling of foF2 Data Based on Satellite Measurements , 2012, AIAI.

[11]  T. Yokoyama,et al.  Temporal change of EIA asymmetry revealed by a beacon receiver network in Southeast Asia , 2015, Earth, Planets and Space.

[12]  Lee-Anne McKinnell,et al.  Neural network-based prediction techniques for global modeling of M(3000)F2 ionospheric parameter , 2007 .

[13]  D. Bilitza,et al.  EQUATORIAL F2-PEAK PARAMETERS,IN THE IRI MODEL , 2003 .

[14]  Dieter Bilitza,et al.  International reference ionosphere—Status 1995/96 , 1997 .

[15]  E. O. Oyeyemi,et al.  Towards the development of a new global foF2 empirical model using neural networks , 2004 .

[16]  Chen Zhou,et al.  Predicting foF2 in the China region using the neural networks improved by the genetic algorithm , 2013 .

[17]  T. Maruyama,et al.  Thailand low and equatorial F2-layer peak electron density and comparison with IRI-2007 model , 2012, Earth, Planets and Space.

[18]  C. C. Lee,et al.  Ionospheric variability at taiwan low latitude station : Comparison between observations and IRI-2001 model , 2008 .

[19]  T. Maruyama,et al.  Low latitude ionosphere-thermosphere dynamics studies with inosonde chain in Southeast Asia , 2007 .

[20]  V. Obridko,et al.  The role of cyclic solar magnetic field variations in the long-term cosmic ray modulation , 2009 .

[21]  D. Bilitza,et al.  International Reference Ionosphere 2007: Improvements and new parameters , 2008 .

[22]  Atef Z. Elsherbeni,et al.  Iterative multiregion technique for large-scale electromagnetic scattering problems: Two-dimensional case , 2005 .

[23]  Weixing Wan,et al.  Using the radial basis function neural network to predict ionospheric critical frequency of F2 layer over Wuhan , 2009 .

[24]  Lee-Anne McKinnell,et al.  The international reference ionosphere today and in the future , 2011 .

[25]  L. McKinnell,et al.  Equatorial predictions from a new neural network based global foF2 model , 2010 .

[26]  L. McKinnell,et al.  On the global model for foF2 using neural networks , 2005 .

[27]  Lee-Anne McKinnell,et al.  Neural network-based ionospheric modelling over the South African region , 2004 .

[28]  Jiankui Shi,et al.  Ionospheric behavior of the F2 peak parameters foF2 and hmF2 at Hainan and comparisons with IRI model predictions , 2007 .

[29]  Y. Shidama,et al.  Using a neural network to make operational forecasts of ionospheric variations and storms at Kokubunji, Japan , 2007 .

[30]  T. Maruyama,et al.  Ionospheric variation at Thailand equatorial latitude station: Comparison between observations and IRI-2001 model predictions , 2010 .