Project Grant 2614589

The National Science Foundation Office of Integrative Activities awarded Rutgers, The State University $449,734 on August 1, 2026, under the Geosciences program (CFDA 47.050) to develop artificial intelligence-enabled, physics-aware methods for forecasting thermospheric density from several hours to several days ahead and to quantify forecast uncertainty and explainability.

The project develops a multi-component framework coupling physics-aware density models with probabilistic predictions of solar and geomagnetic drivers. The first component advances short-term multi-hour probabilistic forecasting by propagating driver uncertainty through nonlinear forecasting mappings to produce calibrated density forecasts and uncertainty estimates. The second component enables long-term multi-day ensemble forecasting using solar imagery and heliospheric inputs to generate probabilistic scenarios for geomagnetic drivers that feed ensembles of physics-aware models. Results will improve understanding of upper-atmosphere variability and support national space-weather preparedness and sustainable space operations.

Work is performed in New Brunswick, New Jersey through July 31, 2029. The project includes workforce development activities training postdoctoral researchers, graduate students, and undergraduate students in artificial intelligence, uncertainty quantification, solar image processing, and thermospheric science. Educational modules, open-source models, datasets, and analysis tools will be made publicly available, and outreach engages high-school teachers and students through hands-on research and seminars.