IBM Corp. says it has come up with a better system for predicting the weather in more isolated parts of the world.

During a presentation at the CES consumer electronics show in Las Vegas today, IBM Chief Executive Ginni Rometty outlined the company’s new Global High-Resolution Atmospheric Forecasting System. The system, she said, would be particularly useful for industries such as agriculture and transportation that rely on accurate, short-term weather forecasts.

GRAF is capable of providing “hyper-accurate” weather forecasts at a resolution of just three kilometers, and it can update itself on an hourly basis, which is a major upgrade for weather forecasts in many parts of the world, Rometty said.

To do this, GRAF will rely in part on data crowdsourced from the barometers of smartphones running the Weather Channel application, which is owned by IBM thanks to its acquisition of The Weather Co. back in 2016. In addition, GRAF will also tap into data from things such as aircraft sensors, giving it access to things such as wind speed and temperature data. This, the company said, would be especially helpful in less developed parts of the world that lack sophisticated meteorological infrastructure.

GRAF’s weather forecasts will be available later in the year via the Weather Channel app, weather.com, the Weather Underground app and wunderground.com. It added that the more accurate forecasts would be able to help airlines minimize disruption from turbulence and help farmers to prepare for adverse weather conditions.

“Today, weather forecasts around the world are not created equal, so we are changing that,” Cameron Clayton, general manager of Watson Media and Weather for IBM, said in a statement. “Weather influences what people do day-to-day and is arguably the most important external swing factor in business performance. As extreme weather becomes more common, our new weather system will ensure every person and organization around the world has access to more accurate, more finely-tuned weather forecasts.”

IBM’s decision to crowdsource data from people’s smartphones is notable because it comes just days after the city of Los Angeles filed a lawsuit against The Weather Co. that alleges it illegally shared and profited from user data collected by its mobile applications. IBM has denied any wrongdoing in that case.

“More data is always better for prediction quality, especially when trying to predict the weather,” said Holger Mueller, principal analyst and vice president of Constellation Research Inc. “That job gets a little easier when you have help from a friend in the sky that can see things real time. The really interesting part is how this new firehose of data will change the prediction models of weather, which is a traditional optimization problem.”

Image: IBM