Google's WeatherNext AI for Weather Prediction for Business

As climate change accelerates weather volatility, so too does the risk to energy infrastructure and supply.

The continuous forecasting and recalibration required to manage these threats are becoming increasingly complex.

In response, technology companies are developing artificial intelligence models that could provide faster, more accurate and lower-cost weather predictions.

According to the World Meteorological Organization (WMO), extreme weather, climate and water-related events caused nearly 12,000 disasters between 1970 and 2021, with reported economic losses reaching US$4.3tn.

These losses have grown substantially over the decades. For the energy sector, climate change is a direct threat to electricity security across both generation and networks.

The IEA found that without timely integration of improved forecasting, power systems could jeopardise up to 15% of wind and solar generation by 2030.

AI forecasting and energy security

AI has the potential to improve weather forecasts by increasing speed, accuracy and resolution.

Organisations worldwide are developing ways to apply AI in forecasting to protect infrastructure and maintain business continuity.

Aardvark Weather, an AI prediction system from University of Cambridge researchers, learns directly from data, making the system simple and flexible.

This approach could allow for rapid adaptation to produce bespoke forecasts for specific industries or locations, such as optimising energy grid balancing or transport routing for maintenance crews.

Nowcasting, which forecasts the immediate hours ahead, can help to enhance disaster preparedness by using real-time information from sources like weather radars and satellites.

The WMO is implementing the AI for Nowcasting Pilot Project (AINPP), which brings together experts from national services, universities and private-sector companies, including Google, Microsoft and Nvidia.

Another example is QubitCast, a quantum-inspired AI platform developed by NASA and Planette, which is capable of predicting extreme weather events months in advance by evaluating numerous atmospheric, oceanic and terrestrial data scenarios.

Next-generation prediction models

Several major technology firms have developed AI-driven platforms for weather prediction.

Google’s WeatherNext is a family of AI models from Google DeepMind and Google Research. Google states that they are faster and more efficient than traditional physics-based weather models.

"WeatherNext will change how businesses use AI for business-critical operations affected by weather, including better planning for retail inventory, logistics disruptions, manufacturing production needs, distribution line maintenance and many other uses," says Carrie Tharp, Vice President, Global Solutions & Industries at Google Cloud.

Carrie explains that by providing advanced forecasting technology, customers can make more informed decisions and ensure stronger business continuity.

"Opening WeatherNext to enterprises expands its applications from the research lab to the real world," says Pete Battaglia, Director of Research for Sustainability at Google DeepMind.

She adds: "It puts companies in the driver's seat to proactively prepare for extreme weather and better serve their communities."

Microsoft’s Aurora is a foundation model developed by Microsoft Research that forecasts a wide range of environmental events at a lower computational cost than traditional methods.

As a foundation model with over a billion parameters, it can be specialised for tasks beyond general weather forecasting, including predicting air pollution and tropical cyclones, even in areas with sparse data.

“We’re not putting in strict rules about how we think variables should interact with each other,” says Megan Stanley, a Senior Researcher at Microsoft Research AI for Science.

She adds: “We’re just giving a large deep-learning model the option to learn whatever is most useful. This is the power of deep learning in these kinds of simulation problems.”

AI-accelerated digital twins

Nvidia’s Earth-2 is a cloud platform for building and running AI-accelerated weather and climate digital twins.

It includes FourCastNet, a global AI forecast model and CorrDiff, a generative model that refines coarse global data into kilometre-scale guidance.

According to Nvidia, CorrDiff is up to 1,000 times faster and 3,000 times more energy efficient than traditional high-resolution methods for a similar task.

Its StormCast model is a generative AI tool for emulating atmospheric dynamics, which could enable reliable weather prediction at a mesoscale, which is critical for disaster planning.

Tom Hamill, Head of Innovation at The Weather Company, told Nvidia: “The production of computationally tractable storm-scale ensemble weather forecasts represents one of the grand challenges of numerical weather prediction.

"StormCast is a notable model that addresses these challenges, and The Weather Company is excited to collaborate with Nvidia on developing, evaluating and potentially using these deep learning forecast models.”

These advancements could offer energy companies new tools for managing assets and mitigating the impacts of severe weather.