NASA Official: A Landsat 8 Clone Would Cost More Than $650 Million

by

WASHINGTON — Building a successor to Landsat 8, the latest in a long-running series of medium-resolution Earth observing satellites, will cost more than the $650 million target Congress asked NASA to hit as part of a 2014 spending bill, Earth Science Division Director Michael Freilich told an agency-chartered advisory panel here May 28.

In responding to Congress, NASA is mulling three possible Landsat 9 architectures, but none fit the constraints lawmakers laid out in the spending bill signed in September, Freilich said.

The legislation calls for NASA to design a Landsat 9 that could “ensure data continuity in an era of increasingly scarce resources with an overall mission cap of approximately $650 million.” Landsat 8, which was turned over to the U.S. Geological Survey in Sioux Falls, South Dakota, in May 2013 to begin service, cost the government about $850 million to build and launch. Five years of operations is expected to push the total to nearly $1 billion. 

NASA’s Goddard Space Flight Center in Greenbelt, Maryland, ran Landsat 8 development, tapping Orbital Sciences Corp. of Dulles, Virginia, for the spacecraft and Ball Aerospace & Technologies Corp. of Boulder, Colorado, to build Landsat 8’s Operational Land Imager camera and a cryocooler for the Goddard-built Thermal Infrared Sensor.

Figuring out how to drive down the collection cost of future Landsat data is one of the key issues the White House has asked NASA to report on by Aug. 15, when the agency will present its imaging strategy for the 20-year period that begins when Landsat 8’s primary mission ends in 2018.

Other than mentioning how high launch costs have diminished the appeal of continuously launching a series of single-instrument spacecraft, Freilich did not provide many hints the strategy NASA intends to propose. 

“Flying multiple spacecraft with one instrument only is probably not the right way,” Freilich said in his May 28 talk to a roomful of senior Earth scientists. When it comes to cost, that approach is “not scoring well right now.”

NASA has been studying ways to ensure the long-term collection of Landsat-grade multispectral and thermal imagery. 

NASA is looking for solutions that can be accomplished for an average annual expenditure of $120 million — the cap the White House set in 2013 when it directed the agency to design what Freilich called a “sustainable and sustained” system of spacecraft to continue the Landsat record after Landsat 8. 

“The problem would be a whole lot easier if we had a lot more money,” Freilich said, adding that the NASA-led team has come up with several “more or less viable architectures, but none of them are perfect.”

One approach NASA could take, and which Freilich has mentioned in interviews before, is to order a block of essentially identical satellites, banking on savings from locking down parts and personnel in a long-term deal. But it is too late to order a copy of Landsat 8 and expect it to cost much less than the one Orbital Sciences shipped out of its Arizona satellite factory in late 2012.

“[U]nless you design basically a block buy system … the second copy that wasn’t planned for costs pretty much the same as the first one,” Freilich said.

For the study the White House ordered, NASA was instructed to keep all options on the table for a Landsat 8 successor, including procuring Landsat-caliber imagery from commercial or international sources. Europe’s soon-to-launch Sentinel 2 series of satellites — which are being designed to carry an optical payload with visible, near-infrared and shortwave infrared sensors to collect images at a 60-meter spatial resolution — might fit the bill, Freilich said. 

Those satellites are designed for just more than seven years of observations from an orbit of 768 kilometers, a bit more than 50 kilometers higher than Landsat 8’s orbit. The first Sentinel 2 would launch in 2015, the second no sooner than two years later, according to the European Space Agency’s website. A single satellite would canvas its observable latitudes once every 10 days. Two satellites would accomplish the feat in five. That compares with one sweep every 16 days from Landsat 8.

However, the Sentinel 2 spacecraft, part of Europe’s planned Copernicus constellation of 15 satellites, lacks Landsat-8’s thermal infrared imaging capability, Freilich said. If NASA decided to rely on the Sentinel 2s for multispectral, medium-resolution images, the agency would have to find some other way to get thermal images, possibly by “doing a U.S. thermal infrared formation flying [spacecraft] with those Sentinels.” 

Although Europe has already started launching Sentinel satellites, one of which launched April 3 from Kourou, French Guiana, “they don’t have the 40-year experience that we have, and there are small but not negligible details that have to be worked out on data exchange,” Freilich said, adding that the U.S. scientists “justifiably … trust NASA. And perhaps they trust the European community a little bit less at this stage.”

However, Freilich hastened to add, “the record shows that when ESA is leading the development, [the mission] pretty much overperforms.”

 

Follow Dan on Twitter:@Leone_SN