Continuous 3D model of ionospheric electron density based on machine learning

Earth's ionosphere represents a complex and dynamic region characterized by increased concentration of charged particles. Changes in ionospheric density can affect the propagation of electromagnetic signals thus disrupting navigation and positioning. The existing models of ionospheric density rarely meet the accuracy requirements due to being either only climatological or using the time and spatial averaging. Here we present, for the first time, a continuous empirical three dimensional (3D) model of electron density at heights 130-900 km. Since the ionosphere is a data rich environment it is essential to use all of the collected observations. We use the radio occultation data from various spacecraft, together with in situ data by CHAMP, GRACE and ROCSAT missions, for over 18 years of measurements. We discuss the application of deep learning to efficiently handle the entire dataset comprising billions of data points. The resulting model gives accurate predictions of electron density in the Earth's ionosphere and yields >90% correlation on the validation data. The model has a wide range of applications for the scientific purposes, space weather monitoring and industrial applications such as positioning and navigation.