Surveillance Satellites Are Not Like Contact Lenses

Nextgov ran a story earlier this month that has me scratching my head. Titled “Pentagon Wants to Use Disposable Satellite Clusters for Intel,” the story raised more questions than it answered, and made some very questionable claims.

From the piece:

“Satellites should be able to deliver imagery within 12 to 96 hours of being launched into orbit and offer persistent coverage, with lapses between captures lasting no longer than 90 minutes, according to the document.

A demo of a constellation composed of 24 satellites is planned for the 2014 to 2015 timeframe. A satellite cluster should cost just $12 million to procure, not including launch and operating expenses.”

Nowhere in the article does it detail how these goals are to be met, or put some context around this idea. While it isn’t a new concept and the mission might be feasible, the economics might be more “DARPA hard” than the technology. Both the Operationally Responsive Space Office (ORS) and the Space Test Program (STP) have proposed similar small satellite programs.  To date, military users have not identified small, rapid response, satellites as a good investment.  Perhaps for this reason, the Administration’s proposed fiscal year 2013 budget call for the elimination of both offices. 

If you dig deeper, you find some pretty aggressive claims regarding the potential economics of this service. When you click on the link in the story to the DAPPA solicitation BAA-12-35, you’ll find the list of program goals, led by this bullet: (you can find them all at this PDF

“Deliver a constellation equivalent number of satellites within 90 days from receipt of order at a specific price point per satellite (based on recurring production costs). The goal is to achieve a complete constellation, including launch, at a fraction of the cost of an existing UAV. For this demonstration, the cost goal is assumed to be $12M for the satellites that make up the constellation, exclusive of launch and operations. (At zero degrees inclination between -10_ and +10_ to support < 90 minutes revisit this translates to, in at least one orbital solution set, 24 satellites.)”

A constellation delivered at a fraction of the cost of a UAV? That could be a challenge.  The average cost of a Predator UAV is approximately $4.5M, far less than the $12M cost projected by DARPA for just the satellite cluster.

Significant costs for the satellite constellation “capability” are incurred in the launch and operations. Granted that when we include 24/7operations, a UAV probably costs 20-30M annually to operate.  However, since it appears that the constellation is only required to last 45 day on orbit, the annual cost goal for delivering the surveillance satellite constellation (feeding a handheld capability called “SeeMe”- Space Enabled Effects for Military Engagements) appears to be significantly HIGHER than a UAV:

http://www.darpa.mil/NewsEvents/Releases/2012/03/12.aspx

Innovation and experimentation have their place, and DARPA certainly has had big successes, including a little something known as the Internet. But in this case, perhaps a program which does not feel the need to justify itself from the on-set as a cost savings over UAVs would allow “DARPA hard” to also be “DARPA believable.”