Finding the proper orbit

For nearly 30 years, the Landsat series of Earth-observation satellites has provided a steady stream of remote-sensing images used by scientists, natural resource managers and others to study changes on the planet's surface. But the Landsat specialty of providing the big picture — deforestation in the Amazon River basin, flooding in the American Midwest, splintering of the Antarctic ice shelf — has made keeping the data flowing a tricky proposition.

Congress wants more private- sector involvement in the Landsat program so taxpayers don't have to foot the entire bill. But there's no commercial market for developing, launching and operating satellites that generate the kind of moderate- resolution data that Landsat users rely on to view wide swaths of the Earth's surface, officials say. Now, an effort under way by NASA and the U.S. Geological Survey to launch a successor to Landsat 7, the latest in the series, aims to address Congress' wishes, industry's misgivings and data users' demands.

Instead of designing its own satellites and sensors and awarding a contract to build them — as it did in the past — under the Landsat Data Continuity Mission (LDCM), the government is only specifying the types of data it needs. Companies will be free to figure out a way to provide that data, which the government will then buy, said Charles Wende, Landsat program executive at NASA's Earth Science Enterprise.

"Basically, our approach is, "You provide the data that meets our specifications,' " Wende said. "How they get there is entirely up to them."

NASA and USGS have been working on a successor to Landsat 7 since that satellite was launched in 1999. The Land Remote Sensing Policy Act of 1992, which authorized Landsat 7, directed managers to study new options for the next mission, with preference given to private-sector initiatives.

Unfortunately, private-sector companies are not interested in building and operating a Landsat 8 because it isn't profitable, Wende said. In the summer of 1999, NASA issued a request for information to gauge industry interest; the response was, " "No, thank you very much,' " Wende said.

Landsats 4 and 5 were handed over to the private sector in the mid-1980s, but commercial success was elusive. Although Landsat 5, launched in 1984 and now operating beyond its design life, can still provide data, USGS officials recently said they would decommission it because Space Imaging Inc., its latest operator, can no longer afford to run it.

There's a commercial market for high-resolution remote-sensing images, such as the 4-meter color and 1-meter black-and-white views provided by Space Imaging's Ikonos satellite, said Ray Byrnes, USGS' liaison for satellite programs. A resolution of 1 meter means the satellite's sensor can detect objects on the Earth's surface that are 1 meter square.

By comparison, Landsat 7 offers lower resolutions ranging from 15 meters to 60 meters. Although such images have important scientific applications, there is virtually no commercial market for them.

"The private sector can't raise the kind of capital needed for a market that doesn't exist," Byrnes said.

The next-best option is the LDCM's data-specification approach, which should give industry more flexibility. Private-sector companies "can come back with creative solutions, and we'll pick the one that is closest to a win-win," he said.

One creative solution might be for a Landsat sensor to hitch a ride on an already-scheduled private satellite, he said. A company could operate two sensors — one purely for commercial imagery and another for Landsat imagery — and the government could agree to be an "anchor customer" that buys most of the Landsat images.

Landsat officials held workshops for data users and industry representatives in January and April to fine-tune the specifications for the next mission. Officials plan to issue a draft request for proposals in late July to invite more comments and then issue a final request in October. The goal is to have a new system in place by the fall of 2005 — about a year after Landsat 7's five-year design life expires.

Continuous coverage is important to researchers, said Jim Plasker, executive director and secretary of the American Society for Photogrammetry and Remote Sensing, because they've spent years building databases documenting changes in crops, vegetation, deserts, coral reefs, cities and more. Landsat 7 provides images of the entire Earth every 16 days.

Not everyone in industry is completely sold on the LDCM concept. Mark Brender, director of Washington operations for Space Imaging, said private satellites already in orbit might be able to provide the needed data.

"The private sector, with many different satellites in orbit, may be able to meet many or most all of the requirements for data continuity at far cheaper cost to taxpayers than building a government satellite," he said. "Why should the taxpayer pay hundreds of millions of dollars when much of the data can be purchased on the open market?"

Using satellites already in orbit is an option NASA would consider, Wende said, as long as a vendor can provide the data users need.

"We're not demanding a new spacecraft," he said. "All [industry has] to do is show that the data meets the specifications. They could put 100,000 carrier pigeons up there with little cameras if that would work."