Skip to main content

As extreme weather devastates communities worldwide, scientists are using modeling and simulation to understand how climate change impacts the frequency and intensity of these events. Although long-term climate projections and models are important, they are less helpful for short-term prediction of extreme weather that may rapidly displace thousands of people or require emergency aid.

An area of climate research known as subseasonal to seasonal prediction fills the gap between the usual daily or weekly weather forecasts and long-term climate projections to provide essential information to help prepare the nation’s mission-critical infrastructure and vulnerable communities worldwide.

Moetasim Ashfaq, a computational climatologist in the Computational Earth Sciences group at Oak Ridge National Laboratory, has studied the Earth’s climate for over 15 years. He began his career as a climate modeler exclusively focused on climate change and its impact 20 to 50 years in the future, but he has recently shifted his focus to analyzing shorter subseasonal to seasonal timescales in which natural climate variability and its global teleconnections — the relationships of distant weather events — are critical to shaping our current climate.

Subseasonal to seasonal predictions rely on the influences exerted by natural forcings, such as ocean temperatures and polar and subtropical jet streams, and how their variations will affect weather in particular regions in the next several weeks to months. Accurately predicting subseasonal weather conditions can help communities better prepare for emergency response in the event of strong storms or unprecedented flooding.

“If we have a subseasonal understanding of processes shaping the Earth’s climate, then we can respond to extreme climate events much better than we have in the past,” said Ashfaq.

Illustration of map labeled with climate and weather variables

This map illustrates the natural climate variability that affects the cold-season climate of the Central Southwest Asian region. Credit: Moetasim Ashfaq/ORNL

This subseasonal to seasonal prediction capability is of particular interest to agencies such as the Air Force Weather, or AFW, program, which relies on accurate predictions to conduct missions safely and effectively around the world. In 2021, the AFW and ORNL launched two HPE Cray EX supercomputers, named Fawbush and Miller, to provide a platform for advanced weather modeling and prediction.

The AFW has expanded its funding to include subseasonal to seasonal research like that conducted by Ashfaq at ORNL, who studies climate patterns in South America, Africa and Asia, as well as the global atmosphere and oceans. Studying multiple regions helps identify the global network of teleconnections established by different forcings and leads to a robust understanding of Earth system processes. It is expected that new research findings through such efforts will improve the skillfulness of prediction models that the AFW can use for areas where it has less information.

The simulations and analytical frameworks that drive this research are run on both the AFW systems and the Oak Ridge Leadership Computing Facility’s IBM AC922 Summit supercomputer to ensure their accuracy and reliability. Highly accurate and near-real-time predictions are critical to the success of AFW and its missions.

“The Air Force Weather Wing relies on real-time weather prediction to keep the United States and our servicepeople safe,” said Kate Evans, director of the Computational Sciences and Engineering Division at ORNL.

The Oak Ridge Leadership Computing Facility is a Department of Energy Office of Science user facility.

UT-Battelle manages ORNL for DOE’s Office of Science, the single largest supporter of basic research in the physical sciences in the United States. The Office of Science is working to address some of the most pressing challenges of our time. For more information, please visit energy.gov/science. -Betsy Sonewald

Betsy Sonewald

Betsy Sonewald is a science writer and communications specialist in the National Center for Computational Sciences division. She highlights news and accomplishments from across the division and its programs.