ºÙºÙÊÓƵ

Finding better ways to handle satellite data as it pours down to Earth is the aim of the GeoStreams research project at ºÙºÙÊÓƵ, funded by an $800,000 grant from the National Science Foundation's Information Technology Research program.

Satellites produce huge amounts of image data every day, and that volume will rise as a new generation of weather satellites are launched over the next decade, said Susan Ustin, director of the California Space Institute at ºÙºÙÊÓƵ and a principal investigator on the grant. Usually, the data is first stored and then processed to pull out information of interest. More effort needs to go into extracting data in real time and processing it on the fly, she said.

Faster processing of satellite data would be especially useful for following fast changing events such as storms, she said.

The ºÙºÙÊÓƵ researchers, led by Ustin and Michael Gertz, a professor of computer science, will develop systems that can look at the data as it arrives, decide what to do with it and pass it on to users according to their needs. The systems will learn what to do from users' demands, Gertz said.

GeoStreams is intended to be scaleable so that potentially thousands of users can make use of the same data stream at the same time.

"What we hope is that through such a system we can provide users with the data they need in an easy to use form," Gertz said.

The project will begin by using data from the National Oceanic and Atmospheric Administration's GOES satellite -- familiar from TV news weather forecasts -- to produce real time maps of humidity and water movement for California. That data can be used for managing irrigation projects in the state.