(TNS) – Researchers at the University of Albany Center for Atmospheric Science Research are working with a Boston tech company to develop new ways to predict wind and extreme weather events much faster and more accurately than current prediction methods.
The company's tomorrow. io provides weather forecasts and data to businesses and government agencies such as NASA and the National Marine Atmospheric Administration using weather observations collected by its proprietary satellites and artificial intelligence platforms.
Tomorrow I've worked with Urbany's atmospheric researchers in the past, but the school and company are now embarking on a new project that says it can provide a much more accurate and practical weather forecast for various industries, using a combination of high-performance computing and real-time weather data collected from both space and ground.
The data from the ground comes from Urbany, which manages the early detection weather observation system in Mesonnet, New York, a network of 127 atmospheric data collection sites located throughout the state.
The partnership-based plan is to combine tomorrow's satellite data, Urbany's Mesonnet System data, and historical weather data to create AI weather forecast models that predict extreme weather events and dangerous winds more quickly and accurately.
The joint release issued by Ualbany and Tomorrow states that the new weather forecasting model is of great value for a variety of industries, including aviation, energy, logistics and emergency management.
New York's Mesonnet Station includes a 10-meter metal tower, the size of a utility pole, located approximately 19 miles from each other, and includes sensors and other weather devices that collect real-time weather condition data. That data is fed into the control room of the Atmospheric Science Research Center, located inside the ETEC building in Urbany, at the Harriman State Office Campus.
Cole Swain, vice president of strategy at Tomorrow.io, believes that the new partnership with Ualbany marks a “significant milestone” as it advances weather forecast modeling.
“The future of weather forecasting is poised to be fundamentally reshaped,” Swain said in a statement. “We are setting stages to dramatically enhance the accuracy and practical utility of weather forecasting and to facilitate the development of next-generation derivative models for critical decision-making.”
Two of the Ulbany researchers working with Tomorrow are Jan Woodcock, director of the operational business of the New York State Excellence Center for Weather and Climate Analysis, and Scanta Bass, an Imperial Innovation Professor who belongs to both the Center for Atmospheric Science Research and the University's Faculty of Environmental and Sustainable Engineering.
Researchers spoke to the Times Union on Thursday about what the weather and climate forecasting tools are important for virtually every industry and government on the planet.
The weather has had a greater impact on the global economy than people imagined, Woodcock said. Extreme weather events can wreaking havoc for crops, transportation, travel and tourism industries and healthcare. Even global finance as an insurance company and reinsurer are working to cover billions of dollars of losses after major weather events.
“Fifteen years ago, 90% of all weather analysis was done by government agencies,” Woodcock said. “What happened (since then) is the weather that has become an important tool used by the industry to manage it more effectively and efficiently, so weather affects more than 35% of the national gross product.
For example, we'll take the power company. In this age of ubiquitous electronic devices, both businesses and consumers cannot function without a reliable supply of on-demand power. As outages will be a major emergency, if major weather events are forecast, the utility will need to prepare a large team of linemen and bucket trucks to prepare to recover power to a location where the storm will land and could cause damage that knocks out the power.
Limitations in current forecasting methods often force utility companies to send large truck crews to two different sites, especially when it comes to wind forecasting, as models cannot accurately identify where the storm will land or where the wind is at its worst, Woodcock explained.
“What they're doing today is for them to send the trucks to both locations,” Woodcock said. “And those that don't show up are called false positives. And each of those trucks costs thousands of dollars a day, which costs millions of dollars.”
According to BASU, AI and high-performance computers are changing weather forecasts very quickly, while also allowing more models to run, so more models in particular offer so much predictability.
“What AI-based systems do, you can run thousands of different scenarios. Not just 10-50, but thousands of different scenarios, and they're really powerful,” says Basu.
The exact scope and timeline of the partnership have not been revealed in detail, but tomorrow. In its release, Io said it was “to quickly deploy, train, and scale” “advanced machine learning models” with the aim of optimizing performance over traditional physics-based predictions.”
©2025 The Times Union (Albany, NY). Distributed by Tribune Content Agency, LLC.
