Follow Datanami:
June 24, 2013

The Big Data Weather Watchers in the Sky

Alex Woodie

A Utah outfit called GeoMetWatch could help usher in an era of dramatically improved weather forecasts if it successfully gets a series of hyperspectral atmospheric sounders into orbit. These big data collectors in the sky will provide a much more detailed view of the movement and composition of the atmosphere, and have the potential to save thousands of lives and billions of dollars in weather-related losses each year.

GeoMetWatch is a spin-off company from Utah State University that hopes to revolutionize weather forecasting with its Sounding & Tracking Observatory for Regional Meteorology (STORM) project. As part of STORM, the company hopes to get six hyperspectral atmospheric sounders into geosynchronous orbit, which will provide a very high resolution view of local weather around the world.

Instead of evaluating one to three atmospheric color bands, as traditional satellite weather sensors do, the STORM collectors will sample 1,800 different colors at 1,800 levels of the atmosphere. “Think of it as doing a CAT scan on the atmosphere,” Forrest N. Fackrell, GeoMetWatch’s Chief Development Officer, told Alex Knapp of Forbes recently.

The sounders will track not only water vapor and temperature, but the presence of atmospheric gasses, wind velocities, and natural and industrial particulates. What’s more, the three-dimensional view of these will be bolstered by the fourth dimension–time. That’s a lot more detailed data than forecasters are working with currently, and has the potential to provide much more detailed forecasts, particular for severe weather events.

So far, GeoMetWatch has signed just one contract to get the first of the six hyperspectral atmospheric sounders into orbit. Asia Satellite Communications Company will include the STORM device (about the size of a refrigerator) on a satellite launch in early 2016. Instead of paying for its own orbital missions, GeoMetWatch plans to rent space on satellites that are already planned for launch.

GeoMetWatch plans to sell the real-time weather data to a variety of domestic and foreign weather agencies and commercial outfits. The data from each sensor is worth from $150 million to $200 million per year, the company says.

That’s a fraction of the losses due to severe storms. The company says that, in the US alone, severe weather kills 2,000 people and causes $7.5 billion in damage every year. GeoMetWatch says its detailed real-time weather data has the potential to increase the accuracy of hurricane landfall estimates by 50 percent, and to provide tornado warnings more than an hour out, as opposed to the amount of time forecasters can provide today, which is typically 10 minutes or less.

GeoMetWatch’s system looks even more promising considering the cutbacks and furloughs currently being implemented at the National Weather Service, and the possibility of losing satellite coverage of the entire Eastern Seaboard. Today’s GOES satellite systems are based on 1970s technology, according to GeoMetWatch.

The National Oceanic and Atmospheric Administration (NOAA) is fully aware of the limitations of its current technology. That’s why the Federal Government spent $400 million to develop the core technology that underlies STORM–called the Geosynchronous Imaging Fourier Transform Spectrometer (GIFTS)–about 10 years ago. After GIFTS was cancelled, the group that developed the technology for the government founded GeoMetWatch to continue its development. Today it’s the only company authorized to sell the technology.

Related Items:

Finding Value in Data Through Life Saving Applications

Can Big Data Help Predict Tornadoes?

IDC Talks Convergence in High Performance Data Analysis