Innovations battle natural calamities

Here at the fall meeting of the "="">Global Earth Observing System of Systems, or GEOSS, a U.S.-led initiative involving 60 nations that are working to integrate data from weather sensors and satellites on a central network to better understand Earth on the whole. The goal is to open up possibilities for global planning, like forecasting winter weather months in advance or predicting the next outbreak of the West Nile virus, for example.

A lightning rod toward this effort is the National LambdaRail Architecture, the first nationwide optical network, or "light pipe," owned and operated by researchers. The network, being built by a consortium of 18 universities and researchers, makes it possible to easily transfer 10 gigabytes of data per second, or roughly 1,000 times more than the bandwidth available on a home high-speed network.

"This is going to enormously change how we make global observations and how we, as scientists, work together," said John Orcutt, deputy director for research at Scripps Institution of Oceanography and president of the AGU.

He was able to recently transfer a 300-megapixel image of a mouse cerebellum from a computer in Copenhagen to one at Scripps in La Jolla, Calif., and then zoom into the image for a clear picture within seconds. "(LambdaRail) allows us to do this without having the data all in one place," he said.

LambdaRail is now in operation, but it's continuing to be developed and extended internationally. Orcutt said he believes the architecture will be introduced commercially in the next four to five years, based on demand from the film industry to send movies this way. Many film companies, he said, wish to transfer films digitally to theaters in order to avoid the cost of delivery.

Orcutt also highlighted Google's vision to provide data on weather systems and local and satellite imagery to the public through Google Earth. Introduced earlier this year, Google Earth is a desktop application that lets people explore cities and landmarks like the Grand Canyon in 3D.

Google Earth CTO Michael Jones introduced the project to scientists at the Center for Earth Observations and Applications inaugural symposium last month. According to Orcutt, Google wants to collect as much data as possible and allow people to add information to it and interact with others using the service.

The search giant recently licensed all of National Geographic's photography archive for use in Google Earth, according to Orcutt. And Google is licensing and gathering new satellite imagery of sea temperatures from NASA and Scripps. Combining all the data into visualizations, people can learn and understand more about their world, he said.

"This is a practical way for the public to get new information about the earth they live on--that's revolutionary," Orcutt said.

"Many more people will become involved in how the world works, and technology has gotten us to the point for these opportunities," he added.

Many of the considerations are cast on the backdrop of a post-Katrina world, in which national disaster preparedness is top of mind for many scientists. Physical oceanographers, biologists and sociologists, among many scientists, are trying to figure out how to better work together to observe and predict disasters. But they're also grasping for new ways of communicating information about Earth and predictable dangers to the public so their warnings aren't dismissed.

If the public is more informed, the thinking goes, projects to protect and rebuild the Mississippi Delta will combine the thoughts and efforts of many interests, including sociologists, politicians, scientists, businesses and residents.

Scientists studying the origins and effects of Hurricane Katrina pointed to the need for visualization tools in order for scientists to gain more knowledge than they would otherwise from simply looking at the data.

Margaret Leinen, assistant director for geosciences at the National Science Foundation, said that the Southern California Earthquake Center recently created a mathematical model for visualizing the effects of a major rupture of the San Andreas Fault that illustrates what areas would be most affected, for example.

"We need to be able to take an image of a hurricane or an earthquake and be able to translate that to decision makers or the public immediately," Leinen said.

"Informatics is the first step we need to take," she added, "to take our research from an observation system to a prediction system."

To this end, scientists and researchers are working to improve sensors and imagery. For example, the University of Oklahoma is spearheading a project called LEAD, which is to change mathematical models on the fly in the event that sensors translate early warnings of a disaster like a hurricane or earthquake. The model changes to project how the Earth and populated areas would be affected, for example.

Researchers at the University of Washington and the University of California, San Diego, are working on building the Laboratory for the Ocean Observatory Knowledge Integration Grid, or LOOKING, which is designed to perfect sensors, such as underwater sensors used to detect when there is a rupture in the Earth's surface. Desired changes to the sensors include infusing them with the ability to collect more data or turn on added sensors once there is an interesting event, such as a rupture.

UCSD's RoadNet is a network of real-time sensors around the globe, or approximately 4,000 sensors of 22 different types, such as TV cameras or coastal radar. Research is going into improving the sensors so that a TV camera can pan, tilt and zoom once it detects an interesting change.

"We're on the bleeding edge of research with this," Orcutt said.