Quantum computing approach for predicting extreme atmospheric events

The National Science Foundation Office of Integrative Activities awarded New Jersey Institute of Technology $266,449 on February 1, 2027, under the Geosciences program (CFDA 47.050) to explore the feasibility of using quantum computing to predict extreme atmospheric events including floods, tsunamis, storms, and turbulent atmospheric bursts.

The project combines quantum information science and geophysical fluid dynamics to investigate whether quantum reservoir computing can provide a computational advantage for predicting extreme events when the target system is described by shallow water equations on the rotating sphere. Classical numerical weather prediction faces prohibitive ensemble costs when capturing rare, high-amplitude events, and data-driven models systematically underpredict them. The project will also train researchers at the intersection of quantum computing, fluid dynamics, and machine learning, addressing workforce development in an emerging field. Better forecasting of heat waves, flash floods, and blocking-driven precipitation extremes carries direct public-safety value.

Work is performed in Newark, New Jersey. The award runs through January 31, 2031. No subawards are planned.

Abstract

This project will explore the feasibility of using quantum computing to help predict extreme atmospheric events including floods, tsunamis, storms and turbulent atmospheric bursts. This project combines quantum information science and geophysical fluid dynamics, two communities that rarely interact. The project will also train an essential quantum-researcher workforce at the intersection of quantum computing, fluid dynamics, and machine learning. Providing better forecasting would have a direct benefit for public-safety value. Prediction of extreme atmospheric events?heat waves, flash floods, blocking-driven precipitation extremes remains challenging: classical numerical weather prediction faces prohibitive ensemble costs when the goal is to capture rare, high-amplitude events, and data- driven models trained on mean-squared-error losses systematically underpredict them. This project would explore whether quantum reservoir computing (QRC) can provide a genuine computational advantage for extreme-event prediction when the target system is described by shallow water equations on the rotating sphere. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the foundation's intellectual merit and broader impacts review criteria.- subawards are not planned for this award.