NOAA preps ground systems for data flood

New satellites will dramatically increase the volume of continuous weather data from which scientists and forecasters must mine useful information.

adrift in a sea of data

NOAA is bracing for a flood of data to come from new satellites. (Stock image)

Over the next four years, two next-generation weather satellite programs will come online under the National Oceanic and Atmospheric Administration, producing an unprecedented deluge of continuous weather data.

While acquisition delays of the Joint Polar Satellite System (JPSS) and Geostationary Operational Environmental Satellite (GOES-R) programs have received the lion's share of public attention because of their $22 billion price tag, a substantial part of the work to ensure the success of both systems is taking place on the ground across the country.

NOAA and component agencies like the National Weather Service are currently in the process of upgrading the vast weather infrastructure and ground services to handle new variations of high-velocity data produced by the satellites and their myriad sensors.

"These satellites are going to produce a firehose of new information to forecasters, but we have to ready our systems to translate all these new capabilities in order to have something that equates to value to the public," said Dr. Mike Johnson, satellite team lead for the NWS' Office of Science and Technology.

"They aren't legacy satellites, we're not replacing existing satellites with similar technology," Johnson said. "This is a big step forward in the science and capabilities we're expecting."

The first GOES-R satellite, scheduled to launch in early 2016, will produce about 40 megabytes of weather data per second, four times more continuous data than existing geostationary satellites.

The satellite's capabilities are magnified by an increased number of channels – from five to 16 – and its ground footprint will increase drastically. In addition, the next-gen GOES-R satellites will be able to scan the continental United States every five minutes as opposed to every 15 minutes like today's models, with the potential to hone in on dangerous weather systems every minute.

When it launches in 2017, JPSS-1's data generation will increase by a similar magnitude, though it will produce less total data – about 6 megabytes per second – than GOES-R.

The sheer volume of data creates delivery problems, Johnson said.

Forecasters at weather terminals require fast-paced data that integrates with radar, and the data flow must be distributed widely to more than 122 weather forecast offices across the country. To cope with the increased data load, NOAA has projects in place to upgrade its Satellite Broadcast Network (SBN), which sorts data streams and directs data to necessary end-users. In particular, SBN's bandwidth requires a substantial upgrade to handle the extra traffic.

In addition, NWS' largest weather facilities, including the National Hurricane Center, are building antennas to receive some satellite information directly. The national weather centers ground networks – they are connected via landlines – will also get revamped. So will data systems maintained by the National Environmental Satellite, Data, and Information Service (NESDIS), which take in additional data produced by the satellites, like volcanic ash data or smoke analysis, for centralized processing.

Finally, Johnson said, developers are working with NWS operations to simulate the pace of data produced by the next-generation weather satellites and how to train forecasters to fully exploit the abundance of data they'll receive. Johnson said his colleagues are working to strategize on these and others issues likely to arise.

"It's quite exciting, these are things that two or three years ago we wouldn't have thought of," Johnson said.