A group of authors consisting of Ivan Heđi, Senior Lecturer at Virovitica University of Applied Sciences; affiliated professors at partner institutions of Virovitica University of Applied Sciences, Prof. Dr. Renato Filjar, Head and Principal Investigator of the Spatial Intelligence Laboratory at the University of Applied Sciences in Krapina, Croatia; Prof. Dr. Jasna Prpić-Oršić from the Faculty of Engineering at the University of Rijeka; and Teodor B Iliev from the University of Ruse, Bulgaria, has published a scientific paper titled “An Ambient Adaptive Global Navigation Satellite System Total Electron Content Predictive Model for Short-Term Rapid Geomagnetic Storm Events.” The paper presents a procedure for developing a model to correct satellite navigation system (GNSS) signal ionospheric delays. The model is based on and adapted to the real environmental conditions surrounding the user’s GNSS positioning device to ensure sustainable quality of position, velocity, and time (PNT) services provided by satellite systems. The development procedure for the model is based on the application of experimental observations and statistical machine-learning methods. The proposed procedure is demonstrated in a critical scenario involving rapidly developing short-term ionospheric storms, where traditional correction methods fail to maintain sustainable GNSS PNT services. During the practical evaluation of the proposed procedure in the statistical computing environment R, three alternative models were developed, based on machine learning and experimental observations collected during rapidly developing short-term ionospheric storms. The models’ performance indicators, including their ability to describe bias and variance, were compared to those of the standard Klobuchar model. The model developed using the Stochastic Gradient Boosting (SGB) method showed the best performance among all the alternative models and the Klobuchar model in the observed critical scenarios. This confirms the applicability of the new model for the observed class of ionospheric disturbance scenarios and their effects on maintaining the quality of GNSS PNT services. The scientific paper was published in the esteemed international journal MDPI Remote Sensing. Remote Sensing is indexed in numerous international scientific databases, including Current Contents (Q1), Web of Science SCIE, and Scopus. The journal’s annual Impact Factor (IF) for 2023 is 4.2, with a five-year IF of 4.9.
