The steady pulse of a storm was once drowned out by the slow processing power of outdated supercomputers. This modernization promises faster updates and more precise warnings for extreme weather events, giving communities more time to prepare and stay safe. NOAA’s new computing strategy will transform how weather predictions are made, ensuring that forecasts remain accurate and timely as extreme weather becomes more frequent due to climate change.
Faster updates, smarter predictions
NOAA is shifting its Weather and Climate Operational Supercomputing System (WCOSS) to Google Cloud. This transition will break free from the rigid structure of traditional on-premises systems. By operating in the cloud, NOAA gains the ability to scale computing power dynamically—especially during critical times like tropical storm season. Scientists will also be able to modify their simulations as real-time data streams in, creating a more responsive forecasting system. The cloud environment allows for rapid updates and seamless adaptations to evolving scientific models, making the forecasting process more efficient.
During intense weather events like hurricanes, the system can quickly ramp up processing to generate more accurate projections of the storm’s path. With faster results, forecasters will be able to issue warnings to coastal communities well before a storm hits, giving people more time to evacuate and take precautions. This improved agility in processing data ensures that weather models stay ahead of fast-moving events, reducing risks for vulnerable populations.
From research to real-time use
According to NOAA Administrator Neil Jacobs, the shift to the cloud will speed up the transition from research to real-world tools, removing major bottlenecks in data processing. The cloud environment provides access to the latest advancements in artificial intelligence (AI) and machine learning, including Google DeepMind. These tools will power new forecasting models that can cut the time needed to generate a global forecast from hours to just seconds. This leap in processing speed will not only enhance NOAA’s forecasting ability but also allow for quicker implementation of new scientific discoveries into operational systems.
The National Weather Service plans to move its Global Forecast System and other key weather models to the cloud beginning in 2027. Full deployment is expected by December 2027. This transition will also include the systems currently in use by NOAA’s 122 local weather forecast offices. These offices will benefit from the same advanced cloud-based tools, ensuring that all levels of NOAA’s forecasting infrastructure are aligned with the latest innovations in technology.
A first for global weather modeling
NOAA will become the first operational numerical weather prediction center in the world to run a complete modeling system in the cloud. This milestone includes creating a computing foundation that can adapt to new scientific discoveries as they emerge. Unlike traditional systems, which require major overhauls to integrate new technologies, the cloud-based infrastructure allows for seamless updates and integration. The flexibility of the cloud environment positions NOAA to stay at the forefront of weather forecasting research and application.
Ken Graham, director of the National Weather Service, described the move as a significant achievement. This new system will allow NOAA to adapt quickly to emerging science, strengthening its ability to fulfill its mission of protecting lives and property across the country. As the transition continues, NOAA aims to ensure that its weather models remain not only the most accurate but also the most resilient in the face of ever-changing weather patterns and technological advancements.