Oct 27 – 31, 2025
Europe/Stockholm timezone

ANN-based software for scaling the ionograms for the needs of operational space weather services

Not scheduled
15m
Thu 30/10: Miklagård - Fri 31/10: Studion

Thu 30/10: Miklagård - Fri 31/10: Studion

Poster APL1 - Space Weather Services and Alerts for End-Users: Bridging Forecasting, Infrastructure, and Communication APL1 - Space Weather Services and Alerts for End-Users: Bridging Forecasting, Infrastructure, and Communication

Speaker

Andriy Zalizovski (Centrum Badań Kosmicznych PAN; Institute of Radio Astronomy, National Academy of Sciences of Ukraine; National Antarctic Scientific Center of Ukraine)

Description

As part of the Pan-European Consortium for Aviation Space Weather User Services (PECASUS) [1], the Space Research Centre of the Polish Academy of Sciences (CBK PAN) traditionally provides global experimental maps of critical frequency of F2 layer (foF2) depression and foF2 depression alerts, calculated in near real-time using data from ionosonde networks. According to the Agreement between CBK PAN and National Antarctic Scientific Center of Ukraine, the data from ionosondes operating at the Akademik Vernadsky Ukrainian Antarctic station (UAS) were implemented into the CBK PAN services.
Parameters of ionosphere such as critical frequencies and virtual heights of the layers are automatically scaled from ionograms. For these purposes the software based on two artificial neural networks (ANNs) with U-Net architecture to identify O- and X- traces of F2 layer was developed. Training the ANNs was carried out on the manually prepared sets of images from ionograms obtained with two ionosondes at the UAS during 2012–2022. Testing of the developed ANNs was carried out on the data obtained during 2023. The absolute differences between the ANN-derived and manually scaled values foF2 were computed to quantify the model’s accuracy. Approximately 50% of the estimates obtained using ANN differ from the manually scaled values by no more than 0.1 MHz.
The software has been running at the UAS server since July 2024, providing high-quality foF2 estimates, and demonstrating good stability. The scaled foF2 values in JSON format are sent to Google Drive every 15 minutes, a few seconds after the end of the ionosonde sounding session and uses for calculating the global maps of foF2 depressions.

  1. Kauristie, K., Andries, J., Beck, P., Berdermann, J., Berghmans, D., Cesaroni, C., De Donder, E., de Patoul, J., Dierckxsens, M., Doornbos, E., Gibbs, M., Hammond, K., Haralambous, H., Harri,A.-M., Henley, E., Kriegel, M., Laitinen, T., Latocha, M., Maneva, Y., Perrone, L., Pica, E., Rodriguez, L., Romano, V., Sabbagh, D., Spogli, L., Stanislawska, I., Tomasik, L., Tshisaphungo, M., van Dam, K., van den Oord, B. Vanlommel, P., Verhulst, T., Wilken, V., Zalizovski, A., and K. Österberg, Space Weather Services for Civil Aviation—Challenges and Solutions, Remote Sensing, 13, 18, https://www.mdpi.com/2072-4292/13/18/3685, 2021.

Primary authors

Andriy Zalizovski (Centrum Badań Kosmicznych PAN; Institute of Radio Astronomy, National Academy of Sciences of Ukraine; National Antarctic Scientific Center of Ukraine) Dr Oleksandr Bogomaz (Institute of Ionosphere, National Technical University “Kharkiv Polytechnic Institute”, Ministry of Education and Science of Ukraine) Mr Artem Reznychenko (Centrum Badań Kosmicznych PAN) Prof. Iwona Stanislawska (Centrum Badań Kosmicznych PAN) Mr Lukasz Tomasik (Centrum Badań Kosmicznych PAN)

Presentation materials

There are no materials yet.