NASA Outsourcing Science

Next week NASA will issue a press release and hold a media event in Denmark to announce the first results from the study of pulsating stars with the Kepler satellite. Kepler was designed to find distant Earth-like planets, but part of the mission is devoted to characterizing the target stars. When NASA holds a press conference, the media usually report on whatever they say -- so this is a great opportunity for a group of scientists who have been working in relative obscurity. Unfortunately, the public may never hear about the most significant results.

Like many NASA projects, the history of the Kepler mission has been riddled with delays and cost overruns. In 2006 NASA approved a 20% increase in the price tag for Kepler, bringing the total cost of the satellite to $550 million and pushing the launch date from 2007 to 2008. A year later the team asked for another $42 million -- but this time the request came to the desk of Dr. Alan Stern, the new associate administrator for NASA's Science Mission Directorate. "Four times they came for more money, and four times we told them no," said Dr. Stern. With threats of NASA turning the project over to new management, it was widely reported that the team cut costs by reducing the mission duration from 4 to 3.5 years and scaling back on preflight tests of the hardware. It is less well known that they also reduced the budget by outsourcing one of the major scientific programs to Europe, where they could externalize the data analysis costs to international partners.

In addition to the primary mission of discovering habitable Earth-like planets around distant stars, the Kepler satellite also has the capability of studying the stars themselves in great detail. The giant digital camera inside the space telescope monitors the brightness of more than 150,000 stars every 30 minutes. This is enough to detect the tiny drops in brightness caused by planets that pass in front of their host stars -- events that typically last for a fraction of a day. But at any given time, the brightness of 512 of those stars is measured 30 times more often. This 1-minute sampling is enough to document the subtle pulsations of starlight caused by continuous "star-quakes" that reveal the properties of the star itself. Thousands of stars have been monitored for one month each during the first year of the mission, and the entire program has been coordinated from headquarters in Denmark at no cost to NASA. Some of the world's leading experts in the study of pulsating stars are based in Europe, so they traded their labor for early access to Kepler data -- effectively extracting a subsidy to NASA from their home countries.

After the first year of the mission, the initial results from these studies were about to be published, so NASA asked the largely-European team to draft a press release. With the help of one of their local media offices (another hidden subsidy), the team pitched a story about Kepler "taking the pulse of distant stars" to learn about their properties -- while developing techniques that could also be used to characterize the host stars of the many planets discovered by the mission. After receiving the draft press release, NASA delayed the target date for the media event by two weeks and asked for additional information from the lead authors of the studies, essentially trashing the draft and starting the process from scratch. This week NASA circulated their "final" version of the press release to the scientists, which largely ignores many of the results and gets the science wrong in what remains -- leaving the scientists to wonder why they agreed to embargo their results for two months.

Kepler Program Scientist Douglas Hudgins was positively giddy about the scientists "quite literally revolutionizing our understanding of stars ... at no cost to the American taxpayer". He's absolutely right about the quality of the work being done by these world-class researchers -- but because of continued mismanagement at NASA, you might never hear about it. I guess you get what you pay for.