The system will allow weather researchers to use personal computers, as well as large-screen projections, to view, interrogate and analyze large observational data sets, including information from radar stations, severe weather detection software, high-resolution weather models, geographic information systems, satellites and aerial photography.
These sources will not only provide weather information, but also data on terrain, building locations and even human activities, such as rush-hour traffic. All of this data will be merged in a platform called the Virtual Geographic Information System (VGIS) previously developed by the project's lead researchers.
Weather researchers will use the visualization system to improve storm detection software used by forecasters. And the National Weather Service may eventually use the system to help decide whether to issue watches and warnings, explains Nick Faust, one of the lead researchers and a principal research scientist at the Georgia Tech Research Institute (GTRI).
"This system will improve weather forecasting in a number of ways," says the other lead researcher, Bill Ribarsky of the Georgia Tech College of Computing. "Ultimately, forecasters will be able to make decisions faster and more precisely. For example, they might see a storm here and make a precise prediction that it's going to affect this community in this way. Forecasters will be able to make more accurate predictions because they will have more information – such as predicting a storm's path based on terrain information and information on human activities. If you have a storm at rush hour, there will be a different effect than if it occurs at night."
The visualization system and associated high-resolution weather models may help forecasters accurately predict general areas of severe weather up to six hours in advance, Ribarsky says. That much lead time could help emergency services personnel know how to respond – for example, whether, when and specifically what areas to evacuate in the case of flooding. Ultimately, predictions integrated with the visualization system could save lives, reduce injuries and save billions of dollars in lost products, equipment and time, Ribarsky adds.
The visualization system is a collaborative project funded by a National Science Foundation grant to Georgia Tech and the University of Oklahoma's Cooperative Institute for Mesoscale Meteorological Studies. Initial prototyping work was accomplished under funding from the Georgia Emergency Management Agency and the Georgia Tech Severe Storms Research Center (SSRC). The National Severe Storms Laboratory (NSSL) in Norman, Okla., is testing and evaluating the system, and also inserting the system's decision-support tools into NSSL severe weather detection software.
photo by Gary Meek
|
|
Severe weather forecasting may be improved by a 3D visualization system under development by researchers in the Georgia Tech College of Computing and Georgia Tech Research Institute. Here, researchers Bill Ribarsky, left, and Zachary Wartell, run the system – which will ultimately work in real time from a PC – and view it on a large-screen projection platform called the Virtual Workbench.
(300-dpi JPEG version - 1.25mb)
|
|
"This system will provide a rich trove of information for analysts and researchers," Ribarsky says. Researchers will not only be able to visualize information in a new way, they will also be able to merge and analyze multiple data sets to study similarities in storm structures in severe weather events. For example, they may merge historical touchdown data from all tornadoes in a certain geographical area to help visually answer the climatological question about whether Georgia has a "tornado alley."
Also, researchers may be able to better understand the impacts of human activities on severe weather. "We know some of the effects of human activities on weather," Ribarsky says. "But perhaps there are some we don't know. So this opens a new research realm."
For example, researchers could use the visualization system integrated with flood extent modeling to allow software predictions of the extent of a flood from a river overflowing its banks into a local area.
"Up to now, all we've been able to get is a flood height (that is, when a river will crest)," Ribarsky explains. "But that doesn't tell you anything about what might be flooded. We will be able to predict the extent of the flood – what parts of the community may be affected, like whether a power plant or chemical plant might be flooded – and convey the information to the public in an easily understood, graphical manner. There are now better ways of estimating the amount of rainfall in a catchment basin. Merging all of these data with the terrain visualization could eventually lead to timely prediction of potential flood damage."
The flood-extent prediction capability of the visualization system is building on collaborative work between Faust and the Atlanta office of the U.S. Geological Survey (USGS). Faust worked with USGS researchers to integrate the USGS flood model for the Flint River near Albany, Ga., into Georgia Tech's VGIS platform. The area surrounding the Flint experienced severe flooding in 1994 and 1998. The collaborative effort helped USGS personnel produce a close estimate of the number of people to evacuate in 1998.
"What we're working on now is a tool that will help officials respond more quickly to severe weather," Faust says. "This visualization system will be very time-sensitive."
An initial version of the visualization system is already receiving north Georgia radar data via the Severe Storms Research Center at Georgia Tech, which gets its feed from the National Weather Service in Peachtree City, Ga. The data is processed by an NSSL severe weather software program that indicates storm signatures, and then automatically transferred to the visualization system for display. Though the visualization system is focusing on north Georgia for now, its structure is scalable – meaning it can be used for local, regional or even national forecasting if data and computing power are available, Ribarsky explains.
courtesy of Zachary Wartell
|
|
The Georgia Tech weather visualization system displays mesocyclone severe weather cells passing over the north Georgia mountains. Mesocyclone cells are features derived from Doppler radar data, which show the position, motion and extent of severe storms.
(300-dpi JPEG version - 442k)
|
|
Researchers are working now to integrate other data into the system. The next step is integration of a high-resolution weather model, which can forecast conditions for areas as small as one to four square kilometers. Researchers expect to complete the project within two years.
"Once we have it all there, we will be able to show for the first time these dynamic volumes of information in this visualization system, basically as the data are received," Ribarsky says. "This has not been done in 3D before in a time-dependent format."
Faust adds that the ability to look at storms in three dimensions in real time will give researchers new insight into the 3D nature of storm development, and that information will result in better severe weather detection software.
In a related research project, Ribarsky and his colleagues are creating a wireless mobile computing system for the new weather visualization system and other applications. In the weather application, doctoral degree student David Krum envisions consumers accessing a personal weather forecast, or "nano forecast," via their personal digital assistant (PDA) or cellular phone. The forecast could be delivered as text and/or a simple graphic.
"The 'nano forecast' is a forecast just for you...." Ribarsky explains. "David Krum is developing techniques to define a user's area of interest (as small as one to four square kilometers). Then the user can get a forecast for when a weather event is going to reach them and from what direction."
Other applications for this wireless "situational visualization" system include finding things like an entrance ramp for a wheelchair-bound person. Emergency personnel could use the system and its dynamic streaming data to find specific locations, or the system could be extended to provide drivers with instant personalized traffic reports.
Researchers realize the commercial potential of the situational visualization system, and some companies have already expressed an interest in the technology, Ribarsky adds.
Working with Ribarsky and Faust on these projects are visiting researcher Chris Shaw, GTRI engineer Tony Wasilewski, and doctoral degree students Krum, Mitch Parry, Zachary Wartell and Richard Zhou.
For more information, contact Nick Faust, Electro-Optics, Environment and Materials Laboratory, Georgia Tech Research Institute, Atlanta, GA, 30332-0841. (Telephone: 404-894-0021)
(E-mail: nick.faust@gtri.gatech.edu);
or Bill Ribarsky, College of Computing, Georgia Tech, Atlanta, GA 30332-0480. (Telephone: 404-894-6148)
(E-mail: ribarsky@cc.gatech.edu)
Contents
Research Horizons
GT Research News
GTRI
Georgia Tech
Send questions and comments regarding these pages to
Webmaster@gtri.gatech.edu
Last updated: Nov. 12, 2001
