iag-comm4-2022-9
https://doi.org/10.5194/iag-comm4-2022-9
2nd Symposium of IAG Commission 4 “Positioning and Applications”
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Neural network model of Electron density in the Topside ionosphere (NET)

Artem Smirnov1,2, Yuri Shprits1,2,3, Fabricio Prol4, Hermann Lühr1, Max Berrendorf5, Irina Zhelavskaya1, and Chao Xiong6
Artem Smirnov et al.
  • 1Space Physics and Space Weather, Geophysics, GFZ German Research Centre for Geosciences, Potsdam, Germany (arsmirnov95@gmail.com)
  • 2Institute of Physics and Astronomy, University of Potsdam, Potsdam, Germany
  • 3EPSS, University of California Los Angeles, Los Angeles, CA, USA
  • 4Navigation and Positioning, National Land Survey of Finland, Kirkonummi, Finland
  • 5Department of Database systems and data mining, Ludwig-Maximilians-Universität München, Munich, Germany
  • 6Department of Space Physics, Electronic Information School, Wuhan University, Wuhan, China

The ionosphere is an ionized part of the upper atmosphere, where the number of electrons in is large enough to affect the propagation of electromagnetic signals, including those of the GNSS systems. Therefore, knowing electron density values in the ionosphere is crucial for both industrial and scientific applications. Here, we employ the radio occultation profiles collected by the CHAMP, GRACE, and COSMIC missions, to model the electron density in the topside ionosphere. We assume a linear decay of scale height with altitude and create a model of 4 parameters, namely the F2-peak density and height (NmF2 and hmF2) and the slope and gradient of scale height in the topside (H0 and dHs/dh). The resulting model (NET) is based on feedforward neural networks and takes as input the geographic and geomagnetic position, the solar flux and geomagnetic indices. The resulting density reconstructions are validated on more than a hundred million in-situ measurements from CHAMP, CNOFS and Swarm satellites, as well as on the GRACE/KBR data, and the developed model is compared to several topside options of the Internation Reference Ionosphere (IRI) model. The NET model yields highly accurate reconstructions of electron density in the topside ionosphere and gives unbiased predictions for all seasonal and solar activity conditions.

How to cite: Smirnov, A., Shprits, Y., Prol, F., Lühr, H., Berrendorf, M., Zhelavskaya, I., and Xiong, C.: Neural network model of Electron density in the Topside ionosphere (NET), 2nd Symposium of IAG Commission 4 “Positioning and Applications”, Potsdam, Germany, 5–8 Sep 2022, iag-comm4-2022-9, https://doi.org/10.5194/iag-comm4-2022-9, 2022.