In the realm of space weather forecasting, a team of researchers from NASA’s Ames Research Center and other institutions has developed a comprehensive dataset designed to improve predictions of the ionosphere, a critical region of Earth’s upper atmosphere. The team, led by Linnea M. Wolniewicz and including Halil S. Kelebek, Simone Mestici, Michael D. Vergalla, Giacomo Acciarini, Bala Poduval, Olga Verkhoglyadova, Madhulika Guhathakurta, Thomas E. Berger, Atılım Güneş Baydin, and Frank Soboczenski, has integrated diverse measurements into a machine learning-ready structure to support advanced forecasting models. Their work was recently detailed in a study published in the journal Space Weather.
The ionosphere plays a pivotal role in global navigation, communication, aviation safety, and satellite operations. However, forecasting its behavior remains challenging due to sparse observations and complex interactions across different layers. The researchers have curated an open-access dataset that combines data from various sources, including the Solar Dynamic Observatory, solar irradiance indices, solar wind parameters, geomagnetic activity indices, and NASA JPL’s Global Ionospheric Maps of Total Electron Content (GIM-TEC). Additionally, they incorporated geospatically sparse data such as Total Electron Content (TEC) derived from the World-Wide GNSS Receiver Network and crowdsourced measurements from Android smartphones.
This novel dataset is temporally and spatially aligned into a single, modular structure that supports both physical and data-driven modeling. The researchers leveraged this dataset to train and benchmark several spatiotemporal machine learning architectures for forecasting vertical TEC under both quiet and geomagnetically active conditions. By doing so, they aim to address gaps in current operational frameworks and enhance the accuracy of ionospheric predictions.
The practical applications of this research for the energy sector are significant. Accurate ionospheric forecasting can improve the reliability of satellite-based communication and navigation systems, which are essential for the operation of renewable energy infrastructure, such as solar and wind farms. Enhanced space weather predictions can also help mitigate potential disruptions to power grids caused by geomagnetic storms, ensuring a more stable and resilient energy supply.
In summary, the researchers have presented an extensive dataset and modeling pipeline that not only supports scientific inquiry into ionospheric dynamics but also offers valuable tools for operational forecasting. This work represents a step forward in addressing the challenges of space weather forecasting and its impact on critical infrastructure, including the energy sector.
This article is based on research available at arXiv.