WASHINGTON — In a move that has longtime Landsat watchers wondering whether history is about to repeat itself, NASA is once again considering cheaper ways to maintain the program’s decade-spanning record of medium-resolution Earth observations.

NASA launched the Landsat Data Continuity Mission (LDCM), the eighth in a line of Earth observation satellites dating back to 1972, in February. Landsat satellites are built by NASA but operated by the U.S. Geological Survey (USGS), which is responsible for processing  and distributing Landsat data to a wide variety of end users. The latest Landsat is set to be turned over to USGS May 1.

Previous attempts to find alternatives to building dedicated Landsat satellites have ended with engineers returning to the drawing board to design another government-owned, free-flying satellite.

“I worked the Landsat issue when I was at the Commerce Department in the early 1990s,” Scott Pace, director of the Space Policy Institute at the George Washington University here, wrote in an April 24 email. “It’s one of those perennial space policy issues that every administration faces.”

Pace, who went on to work for NASA during then-President George W. Bush’s drive to return astronauts to the Moon, said he doubted hosting Landsat sensors aboard other spacecraft would work as an alternative to building a ninth Landsat satellite, but he said a data-buying arrangement, with either foreign governments or commercial satellite-image providers, might.

“The difficult issues will be in firmly fixing the data continuity requirements early, ensuring cost-schedule confidence levels are given high weighting in the evaluations, and having a realistic government back up option should the data buy option falter,” Pace wrote.

For 2014, NASA requested $30 million to study a follow-on to the Landsat Data Continuity Mission. The results of the study will be part of the agency’s 2015 budget request, which is expected next February. NASA hopes to discover a feasible alternative to building and flying another dedicated satellite.

The LDCM satellite has two main instruments: a $188 million medium-resolution Operational Land Imager, built by Ball Aerospace & Technologies Corp. of Boulder, Colo., and the $160 million Thermal Infrared Sensor, which was built at the Goddard Space Flight Center in Greenbelt, Md. The spacecraft bus was built under a $194 million contract Dulles, Va.-based Orbital Sciences Corp. inherited when it bought General Dynamics’ Gilbert, Ariz.,-based satellite business in 2010.

The final tab for Landsat 8, as the latest in the series will be known after May 1, is expected to be about $930 million, including development, launch and five years of operations, according to the U.S. Government Accountability Office.

So cost will be a factor as NASA considers alternatives, including the use of “stand-alone new instruments and satellites, as well as potential international partnerships,” according to the plan the agency laid out in its 2014 budget request, which was released April 10.

“Instead of another $900 million satellite, we’re going to use this seed funding in 2014 to look across agencies, in the U.S. as well as outside, as well as private sector capabilities,” NASA Deputy Administrator Lori Garver said during an April 22 Earth Day tour of the Goddard Space Flight Center, which oversaw LDCM’s construction. “It is about the data coming in a continuous fashion.”

A longtime Landsat scientist who shadowed Garver during the Earth Day tour acknowledged the high cost of the last Landsat satellite, but balked at the idea of starting discussions of a follow-on based on cost.

“I agree that Landsat Data Continuity Mission, [at] $850 million for us to launch, was not inexpensive,” James Irons, the mission’s project scientist, said to members of the press April 22. But, he said, “I would like to begin the conversation with requirements rather than with costs.”

For the basic requirements, medium-resolution images of wide swaths of land, there are some international options NASA might tap into to supplement its Landsat archives, Irons said. India’s Resourcesat-2, which launched in 2011, captures Landsat-like data, as will a pair of Sentinel-2 satellites planned by the European Space Agency, the first of which is set to launch in 2014.

But “international cooperation is a difficult road,” Irons said. And if international cooperation is difficult, commercial image procurement is nigh impossible for this program, in Irons’ view.

“We tried to buy data initially for the [Landsat Data Continuity Mission], and that strategy was abandoned,” Irons said. “It culminated with the release of a request for proposal in 2003, and the proposals that were received … were not accepted by NASA.”

“The final bidder on that, Resource21, packed it in at the last minute because NASA headquarters was unwilling to pay all the freight for them to do a data buy,” Samuel Goward, a professor at the University of Maryland’s Department of Geographical Sciences in College Park, Md., and one-time Landsat science team leader, said in an interview.

Back in 2003, NASA thought the commercial sector could provide the data needed for the Landsat Data Continuity Mission. When the agency solicited bids that year, Resource21 and DigitalGlobe — which has since become the sole provider of commercial satellite images to the U.S. National Geospatial-Intelligence Agency — were both expected to submit proposals.

Only Resource21 did, and NASA decided the company’s $500 million asking price was too steep. The agency withdrew its request for proposals and moved to host Landsat instruments aboard a civil-military polar-orbiting weather satellite. However, by the time the National Polar-orbiting Operational Environmental Satellite System was canceled in 2010, the White House had decided to have NASA build a free-flying replacement for the aging Landsat 7.

Asked April 22 whether a data-buying effort for Landsat could be successful today, Irons replied simply, “No.”

Goward, while also not sanguine about the prospects of finding a commercial Landsat vendor, noted there has been at least one “moderately successful data-buy program” where a NASA science mission used data from a commercially hosted instrument: the Sea-viewing Wide Field-of-view Sensor that launched in 1997 aboard GeoEye’s OrbView-2 satellite. That sensor went dark in 2010.

Near the top of Goward’s nice-to-have list for a Landsat 8 follow-on is a second satellite to follow the primary Landsat imager across the globe. Currently, Landsat satellites pass over the same spot of ground every 16 days. Putting a second satellite in the train would produce images every 8 days, Goward said.

“The easiest way to do that is to wait for Sentinel 2 to launch,” Goward said. “But that’s under the assumption that the Europeans are going to give free access to the data the way the U.S. gives free access to Landsat data.”

One industry source familiar with NASA’s last effort to find an alternative to dedicated Landsat spacecraft said that judging by the language in the 2014 budget request, the agency probably was not interested in an entirely commercial solution — despite the possibility that such an arrangement could be cheaper than a dedicated Landsat 9.

“A commercial company could provide a data services contract to the U.S. government … that would allow NASA to sustain the Landsat data stream at a cost far less than what would be incurred with full NASA involvement,” this source said, citing the expense of civil service labor during sensor and spacecraft development as a primary cost driver.

Dan Leone is the NASA reporter for SpaceNews, where he also covers other civilian-run U.S. government space programs and a growing number of entrepreneurial space companies. He joined SpaceNews in 2011.Dan earned a bachelor's degree in public communications...