SAN FRANCISCO — Five days before approximately 200 tornadoes swept through six states in the Southeastern United States between April 25 and April 28, local meteorologists were alerted to the possibility of unusually severe weather by National Oceanic and Atmospheric Administration (NOAA) satellite data.
“Forecasters might not have known the exact county and city where the storms would hit, but they knew the state or local area five days in advance,” said Tom Renkevens, deputy chief of NOAA’s Satellite Products and Services Division in Camp Springs, Md.
Before the storms showed up on ground-based radar, it was difficult for meteorologists to pinpoint their location. Nevertheless, several days ahead of time, NOAA’s polar-orbiting satellites detected the varying atmospheric temperature and moisture levels that signaled the threat of severe thunderstorms and the possibility that those storms could produce tornadoes. That information was used in NOAA’s weather forecasting models and provided to meteorologists, NOAA officials said.
“Five days before the storms, we had a forecast showing a high risk for significant severe weather,” said Greg Carbin, warning coordination meteorologist for the NOAA National Weather Service Storm Prediction Center in Norman, Okla. Ten to 12 hours before the storm, forecasters equipped with additional satellite imagery showing the location of the storms began warning residents of the possibility that violent tornadoes would touch down in their local communities, creating damage along an extended path, Carbin added.
To provide long-range storm warnings, NOAA employs five Polar Operational Environment Satellites (POES) each equipped with sensors to gather information on atmospheric temperature, moisture, wind speed and direction at different altitudes. Those data are important because the distribution of temperature and moisture as well as shifts in wind direction aid in the prediction of severe weather, Renkevens said.
As the recent storms moved closer to the United States, NOAA officials relied on the instruments carried aboard the agency’s two Geostationary Operational Environmental Satellites (GOES) to monitor rapidly changing weather conditions. NOAA officials typically use the GOES imagers to gather observations of the entire United States every 15 minutes. When severe weather strikes as it did in late April, the schedule for the satellite’s imagers is modified to provide data every 7.5 minutes, Renkevens said.
In addition to the contributions of satellite-derived data, NOAA officials attributed recent improvements in long-range forecasting and the accuracy of severe weather warnings to dramatic improvements in computer models. “Models have gotten much better at making accurate forecasts for severe weather or snow farther in advance of storms,” Renkevens said.
The Storm Prediction Center uses antennas to obtain data directly from the NOAA satellites less than one minute after the information is captured by onboard sensors, Renkevens said. The Storm Prediction Center and National Weather Service offices across the country also gain access to satellite imagery through the agency’s computer network. NOAA computers in Suitland, Md., feed millions of daily observations drawn from space-based sensors into the high-resolution computer models that produce weather forecasts, Renkevens said.
Once tornadoes are spotted in local communities, Doppler radars track their movement. By combining information from radars, satellites and local observations with computer models, meteorologists were able to provide residents of the Southeastern states struck by the late April storms with an average warning time of 27 minutes. Still, those storms killed an estimated 311 people, Carbin said.
NOAA officials said severe storm warnings are likely to become far more timely and accurate when meteorologists gain access to data captured by NOAA’s next generation of geostationary satellites. The first of those satellites, GOES-R, is scheduled to launch in December 2015, said Tim Schmit, a meteorologist at NOAA’s Center for Satellite Applications and Research in Madison, Wis. NOAA officials plan to begin using those data to produce operational weather models in early 2017 after engineers confirm that the spacecraft and its instruments are working properly and scientists confirm the accuracy of the data obtained, Schmit said.
Satellites currently flying in the GOES constellation obtain imagery in five spectral bands in contrast to the GOES-R spacecraft, which will carry the Advanced Baseline Imager built by ITT Corp. of Fort Wayne, Ind., to gather data in 16 spectral bands. With information provided by those additional spectral bands, meteorologists will be able to gain a much clearer picture of cloud structure and composition, including information on whether moisture in clouds is cooling rapidly, which is one indicator of severe weather, Schmit said.
The imager also will be capable of capturing a complete image of the United States every five minutes. During severe weather, GOES-R will be able to scan an area measuring 1,000 kilometers by 1,000 kilometers every 30 seconds or two different areas of that size every minute, Schmit said.
In 2008, NASA — acting on NOAA’s behalf — awarded a $1.9 billion contract to Denver-based Lockheed Martin Space Systems to build two GOES-R spacecraft. The contract includes options for two additional GOES-R satellites. In addition to improving weather forecasts, GOES-R features instruments that will aid scientists and emergency managers in monitoring sea surface temperatures, environmental aerosols, volcanic ash plumes and lightning, NOAA officials said.
