CIRES Scientists Awarded $5.3M for Space Weather Research
NASA and the National Science Foundation have awarded two CU Boulder space weather scientists more than $5M to lay the groundwork for faster and more robust space weather forecasts. Both projects are led by CIRES scientists working with the NOAA Space Weather Prediction Center.
Changing conditions on the Sun and in space—broadly called "space weather"—can affect various technologies on Earth, blocking radio communications, damaging power grids, and diminishing navigation system accuracy. Accurate forecasting of energetic events on the Sun and in the near-Earth space environment is critical for national security and society’s well-being.
Funded by NASA, CIRES scientist Enrico Camporeale will apply probabilistic modeling, estimates of uncertainty, and machine learning to space weather forecasts to improve their accuracy.
Funded by the National Science Foundation (NSF), CIRES scientist Tzu-Wei Fang will investigate upper atmosphere conditions leading to disturbances that could interrupt GNSS and other radio signals.
NASA and NSF are funding these projects as part of their joint Space Weather with Quantified Uncertainties program, with the goal of bringing together teams from across scientific disciplines to advance the latest statistical analysis and high-performance computing methods within the space weather modeling field. Together, NSF and NASA are investing over $17M into six three-year awards.
“These awards will make sure that recent, extraordinary advances in computer modeling and data assimilation techniques are applied to critical questions in space weather research,” said Mangala Sharma, NSF program officer in the Division of Atmospheric and Geospace Sciences.
Camporeale heads up the project “Ensemble Learning for Accurate and Reliable Uncertainty Quantification,” awarded $2.9M by NASA. The project’s goal is to combine a small number of high-fidelity—low error, but computationally expensive—runs from physics-based models with a large number of possibly less accurate, but faster, runs from machine learning models. This approach will allow the researchers to accurately estimate the uncertainties associated with their predictions.
Aiming to improve current space weather forecasts, Camporeale and his colleagues will put their methodology to the test, using it to predict solar wind under ambient conditions and during coronal mass ejections; electron fluxes at geostationary orbit; and Earth’s magnetic field.
“This project is very interdisciplinary, involving space weather, machine learning, and uncertainty quantification, so it will benefit the broader field of machine learning applied to physical sciences,” Camporeale said.
Fang heads up the project “Forecasting Small-scale Plasma Structures in the Earth’s Ionosphere-Thermosphere System,” awarded $2.4M by NSF. Space weather is caused by disturbances in Earth’s ionosphere, part of Earth's upper atmosphere that reflects and modifies radio waves used for communication and navigation. These disturbances can disrupt transmission of Global Navigation Satellite System signals critical for precise positioning, navigation and timing, and can influence transmission of radio waves important for land-satellite communication.
Fang and her colleagues will model the conditions that lead to disturbances in the ionosphere and use ground- and satellite-based observations to validate the simulated ionosphere-thermosphere conditions. The researchers’ goal is to understand how perturbations in the upper atmosphere system lead to those disturbances.
“Our goal is to advance fundamental research underpinning space weather forecasts. Our research will help us better forecast disturbances in the ionosphere and their impact on the satellite signals,” Fang said.
This news story was adapted from a press release by the National Science Foundation. Read the full release here.