David Roy of SDSU heads NASA grant satellite data project

A major grant from NASA will help South Dakota State University scientists make satellite data easier to use and access via the Internet.

The National Aeronautics and Space Administration has awarded $3.29 million to a five-year project led by professor David Roy of SDSU’s Geographic Information Science Center of Excellence.

Roy said the project is a collaboration with the U.S. Geological Survey, or USGS, through its Center for Earth Resources Observations and Science near Sioux Falls. The USGS center is the main federal repository for satellite images including those taken from Landsat. Since 1972, Landsat satellites have been sensing dedicated images of the Earth — now the longest such record in existence.

“This project will provide the USGS with a state-of-the-art strategy for creating the land monitoring data sets needed by the nation's resource managers and an exciting opportunity for evaluating the next generation of Landsat processing and delivery systems,” said Tom Loveland, the USGS Landsat science team leader. “The project is timely because as of this year, Landsat images became available free of charge.”

Roy said the proposal is to take Landsat observations every 16 days for all the contiguous United States and Alaska for a seven-year period and process those data so that they’re available to the user community over the Internet in a seamless manner. Researchers also intend to characterize the land cover from the data.

“Right now if you’re in a high school or you’re in a geography department of a university and you want to use Landsat data, you have to be an expert in processing and accessing the data,” Roy said. “The point here is that the user community really wants to be able to obtain processed Landsat data more simply.”

The project also sets out to fill in gaps in the data. Since 2003, Roy said, there has been a problem with the Landsat sensor so that it has been unable to record the data for about 22 percent of each image. In addition, an average of about 35 percent of the Landsat data is obscured by clouds. Those gaps present ongoing challenges for users.

“We need a way of getting rid of the gaps,” Roy said.

The SDSU project will do that by using data from another satellite system called MODIS. Roy worked extensively with that system in a previous position at the NASA Goddard Space Flight Center in Maryland.

Integrating the data from the two satellite systems can be done but it will be complicated. Scientists will spend the first half of the five-year period developing a prototype at SDSU’s GIS Center of Excellence. Then they’ll move the system to the USGS and put it into operation.

“This work over the next five years will be one of the benchmarking exercises for perhaps working out how to do processing and distribution for the next generation of Landsat,” Roy said.

Tom Loveland and SDSU professor Matthew Hansen are co-investigators on the project.

“This project has the potential to fundamentally change the way satellite data are accessed and used,” Hansen said. “Our experience in processing and characterizing Landsat imagery will enhance the utility of the data for a whole range of users. This is a very exciting project.”

[via]