A novel neural network model of Earth’s topside ionosphere
Artem Smirnov, Yuri Shprits, Fabricio S. Prol, H. Lühr, Max Berrendorf, Irina Zhelavskaya, Chao Xiong
Abstract
The Earth's ionosphere affects the propagation of signals from the Global Navigation Satellite Systems (GNSS). Due to the non-uniform coverage of available observations and complicated dynamics of the region, developing accurate models of the ionosphere has been a long-standing challenge. Here, we present a Neural network-based model of Electron density in the Topside ionosphere (NET), which is constructed using 19 years of GNSS radio occultation data. The NET model is tested against in situ measurements from several missions and shows excellent agreement with the observations, outperforming the state-of-the-art International Reference Ionosphere (IRI) model by up to an order of magnitude, especially at 100-200 km above the F2-layer peak. This study provides a paradigm shift in ionospheric research, by demonstrating that ionospheric densities can be reconstructed with very high fidelity. The NET model depicts the effects of numerous physical processes governing the topside dynamics and can have wide applications in ionospheric research.