Failure Of Landsat 6 Leaves Many Researchers In Limbo

On its launch this past October 5, the Landsat 6 remote- sensing satellite crashed into the Indian Ocean--a sourly disappointing turn of events for researchers in several scientific fields. These scientists have grown increasingly dependent on the space device, which scans specific electromagnetic wavelengths from Earth, to supply them with unique data on the ever-changing planet. For these researchers, this latest setback is representative of the roller-coaster history of the two-decade-old

Myrna Watanabe
Dec 12, 1993
On its launch this past October 5, the Landsat 6 remote- sensing satellite crashed into the Indian Ocean--a sourly disappointing turn of events for researchers in several scientific fields. These scientists have grown increasingly dependent on the space device, which scans specific electromagnetic wavelengths from Earth, to supply them with unique data on the ever-changing planet.

For these researchers, this latest setback is representative of the roller-coaster history of the two-decade-old Landsat satellite program, and their experience with it. It is a history marked, on one hand, by the spectacular research applicability of the satellite, far exceeding its initial expectations, and, on the other, by administrative and political decisions that eventually priced the data it collected out of the reach of most researchers, forcing them to abandon it as a viable tool.

The first of the Landsat series of satellites was launched by the U.S. Department of the Interior in 1972. The program was a direct outgrowth of the use of hand-held Hasselblad cameras by the astronauts in the Mercury and Gemini space programs to photograph the Earth from space, according to Paul Lowman, a geophysicist with the National Aeronautics and Space Administration's Goddard Space Flight Center in Greenbelt, Md. These space photos enthralled geologists.

The Landsat program was a concerted effort to monitor the Earth from space on a regular and repetitive basis with sensors that detect radiation in select wavelengths, providing information that geologists could use to study landforms, land movement, and the Earth's geological history. These satellites repeat coverage of the same area approximately once every 16 days. The digital data obtained can be analyzed by computer and can be used to make black- and-white or multicolored images of the Earth.

Landsat 2 was launched in 1975, Landsat 3 in 1978, Landsat 4 in 1982, and Landsat 5 in 1984. Landsat 4 and 5 have improved sensors and produce images with increased resolution, but the resolution is inferior to that of the French-European Systme Probatoire pour l'Observation de la Terre (SPOT) satellite, an alternative available to researchers. The average lifespan of Landsat 1 through 3 was six years. Landsat 4 and 5 are still aloft, but Landsat 4 has been shut down because of communication difficulties and Landsat 5's transmissions are weak.

Scientists expect that Landsat 7, to be launched in 1998, will carry sensors with greater resolution equivalent to SPOT, but Tony Janetos of NASA in Washington, D.C., explains that if the launch is moved up, this improvement may have to be scrapped.


A recently passed federal law revived the hopes of many of these scientists that they might again use the satellite in their studies, but the October crash and federal budgetary problems have left those hopes, at the very least, in limbo.

The first Landsat satellite, launched in 1972, was expected to be used primarily to aid geologists in their monitoring of the Earth's surface from space on a repetitive basis with sophisticated sensors.

It wasn't long, however, before scientists in many other fields--geographers, ecologists, biologists, environmental scientists, and others --realized that remote sensing of the Earth's surface could provide them with invaluable information. Use of the medium blossomed; suddenly, Landsat imagery was conveying startling data--in the form of striking and informative color and black-and-white images-- about crop yields, irrigation, ecological recovery from brush fires and volcanoes, siltation and diversion of rivers, and other phenomena.

And the price was right. A data tape produced from the satellite could be obtained for $250, a black-and-white print for $10.

But 12 years later, in 1984, the Reagan administration moved the Landsat program from governmental auspices to the private sector. Run by the National Oceanic and Atmospheric Administration (NOAA) under the Department of Commerce in the early 1980s, the satellite's operation and development were put under the aegis of a private company, Lanham, Md.- based Earth Observation Satellite Co. (EOSAT), a joint venture of Camden, N.J.-based RCA Corp. and Hughes Aircraft Co., Los Angeles. Almost immediately, prices for the satellite-generated imagery skyrocketed: The computer- compatible tape now cost $4,400; the price of the $10 black- and-white print soared to $2,700.

Some investigators made do with pre-EOSAT imagery to continue their studies, while others reduced their research, exploited expensive and non-comparable imagery from European satellites, or began making the best of imagery obtained from conventional aircraft.

Understandably, they had reason to be hopeful when, in October 1992--after much congressional lobbying by them and sympathetic politicians--a law was passed that will eventually see the return of Landsat to governmental supervision and affordability. The Land Remote Sensing Policy Act of 1992, signed by President George Bush, requires that Landsat be turned over to the National Aeronautics and Space Administration and the Department of Defense from EOSAT. The law sets up a procedure that will make data available at cost to specific researchers. The process of buying back Landsat has begun, but NASA does not yet have the budget to buy its share; however, it has bought some Landsat data from EOSAT and has made them available to researchers.

But then came the Landsat 6 debacle, a loss of valuable resources that many were quick to blame on the ineptitude of NOAA, which handled the launch, and EOSAT, which had built the satellite. Once again, the scientists whose work depended on this remarkable device are concerned about the prospects of continuing their studies. The next generation of Landsat satellite, Landsat 7, is scheduled to be launched in 1998, although plans to speed up the process in the wake of the Landsat 6 crash have been discussed. In the meantime, scientists must wait or find alternatives.

Many Applications There is no way of monitoring large-scale environmental degradation and global change except through remote sensing. For example, the University of New Hampshire's David Skole and his colleague Compton Tucker of NASA's Goddard Space Flight Center in Greenbelt, Md., used the data to determine rain forest loss in Brazilian Amazonia. Their results (Science, 260:1905-10, 1993) showed much less of a loss than environmentalists expected.

They based their assessment on a massive amount of Landsat- generated information obtained from the Brazilian space agency, which has a monitoring station to receive data directly from satellites. These stations, in countries throughout the world, pay EOSAT for access--the privilege of turning on the satellite and collecting and processing the data themselves. Skole and Tucker used approximately 250 Brazilian scenes at $250 per scene, Skole says. He calculates that the same scenes, if available from EOSAT, would have cost $1.1 million.

Victor Klemas, a professor of marine studies and director of the Center for Remote Sensing at the University of Delaware in Newark, uses remote sensing imagery to assess erosion and pollution. This is important not only to assess water quality, but also to prepare for possible oil spills in the region, according to Klemas. Klemas has purchased data from EOSAT from grant money for some of his projects. "All in all, the research has suffered because of the high price structure," he says.

He sees himself in a bind as a result of the loss of Landsat 6. "We all counted on Landsat 6 data for continuity," he says, referring to the need for repetitive coverages and images in similar formats to those taken by Landsat before.

Another use of Landsat remote sensing data is for monitoring resources. According to Janine Stenback, a remote sensing specialist with the State of California's Department of Forestry and Fire Protection in Sacramento, her state specifically is looking at the conversion of wild lands to urban lands, monitoring threatened and endangered habitats, and mapping and monitoring wetlands.

California, however, cannot afford EOSAT's prices and was looking forward to receiving relief under the new Landsat 6 pricing structure. According to Stenback, California has been relying on aircraft-acquired data--which are in similar wavelengths as the data from Landsat--from NASA's Ames Research Center in Mountain Home, Calif., for emergency response situations, such as quickly locating and assessing the scope of brush fires.

This coverage is both frequent and rapid, and, in emergency situations, the data were provided free of charge. But these data are limited in use, Stenback says. It is hard, for example, to put the images together to form a continuous mosaic; and images may differ, depending upon the altitude of the aircraft.

Donald Rundquist, a geographer who is a professor of conservation at the University of Nebraska in Lincoln, says that he relies on Landsat remote sensing imagery for many studies, most of which are related to the agricultural emphasis of his school. These include irrigation monitoring; crop classification, identification, and mapping; and water- quality assessment. His group had to cut back on the imagery after EOSAT raised the prices. Now, he is considering alternatives.

Alta Walker, a geologist with the United States Geological Survey in Reston, Va., used contemporary Landsat imagery in the early 1980s to determine the extent of remaining natural habitat for the endangered Chinese alligator. This required imagery from several different years and seasons, and would be prohibitively expensive now. Walker now limits her research to geology and geological history, which do not require time-sensitive images, trading old Landsat tapes with other researchers and the government.

Meanwhile, Fred Koontz, curator of mammals at the Bronx Zoo/Wildlife Conservation Park in New York, has used satellite tracking systems to follow movements of radio- collared elephants in Africa. He had hoped to use Landsat remote sensing imagery for on-site field research and wildlife conservation programs, but under EOSAT, such imagery has been out of financial reach. And now he foresees a "data gap" until Landsat 7's proposed 1998 launch.

Bridging The Gap Since EOSAT took over the Landsat program, consumption of Landsat data by academics dropped precipitously. In 1976, researchers purchased 34,000 scenes, an all-time high; in 1990, only 450 scenes were purchased, says Koontz. Even the scientific literature suffered. Geophysicist Paul Lowman of the Goddard Space Flight Center says that very little has been published on use of remote sensing imagery in the last several years.

The new pricing structure that will be worked out as a result of the 1992 law was expected to lead to immediate increased demand for Landsat 6 data. Given the crash and ongoing negotiations to repurchase the program, this will not happen.

How can the many researchers hope to continue their work? Right now, there are only two barely functioning Landsat satellites: Landsat 4, which has been shut down because of communication difficulties, and Landsat 5. Both of these are long overdue for replacement, and both of them can fail at any moment.

Although some existing Landsat data will be provided at cost to certain researchers and organizations under the new law, current data will be available only as long as the two satellites last.

Archival data, now maintained on tapes in storage centers by NOAA, have the disadvantage of being dated; moreover, they tend to not be in the best of shape. According to Skole, much previously obtained material is located in coun- tries outside the U.S. that have their own tracking stations. In fact, Skole points out, no one knows exactly what data are available worldwide.

Alternatives Those who can afford to are considering alternatives, but nothing seems to be ideal. California is looking into using data from the French-European Systme Probatoire pour l'Observation de la Terre (SPOT) satellite. But SPOT data are not cheap, and, California's Stenback says, the differences in resolution and format may make it difficult to compare Landsat and SPOT imagery of the same area.

Rundquist is considering using imagery from the Japanese JERS-1 satellite for work on changes in the aquifer in four western lakes in Nebraska. These data also are expensive-- about $1,000 for a computer-compatible tape--but significantly cheaper than EOSAT's data. New Hampshire's Skole questions the use of Japanese imagery. "It's of dubious quality," he says.

The best solution seems to be to move up the launch of Landsat 7. Landsat 7 is not a panacea; its technology, according to Nebraska's Rundquist, is dated. But it still will provide affordable data to U.S. researchers.

Researchers such as Koontz are concerned that national budgetary problems may further hinder the project. Koontz adds, however, that President Clinton and Vice President Gore support the Landsat program. "The government is going to have to step in and follow through on some of these ideas, or the U.S. will lose its leadership position," Koontz states.

Myrna E. Watanabe is a biotechnology consultant based in Yonkers, N.Y.