An NGA-NRO partnership could lead to cueing of commercial imagery satellites

Perhaps no one in U.S. government is more excited about the ongoing boom in commercial satellite imagery than intelligence officials.

Last month, two U.S. intelligence agencies — the National Geospatial-Intelligence Agency and the National Reconnaissance Office — announced a new collaboration to help their organizations exploit commercial satellite imagery. This effort, known as the Commercial GEOINT Activity, is slated to kick off by the end of September and is the latest in a series of initiatives to better take advantage of new commercial offerings.

The wave of new imagery providers, IC officials believe, will translate into better and faster intelligence.

“We’re not standing up a large organization, but the idea is that it’s advisory,” said Orrin Mills, NGA’s lead official on the project. “It’s advising the parent agencies on what the best capabilities against the requirement are.”

The NRO, which builds and operates the United States’ spy satellites, is responsible for collecting imagery while the NGA is responsible for processing those images into actionable intelligence. The new collaboration will help synchronize NRO and NGA commercial imagery investments.

The move comes after the intelligence community has spent the better part of the last two years studying how to best work with small satellite companies, many of which have little experience negotiating the bureaucratic hurdles selling to the federal government entails.

How will this initiative be different? NGA and NRO officials explained the practical implications of the Commercial GEOINT Activity in an interview with SpaceNews’ Mike Gruss.


SpaceNews: Both the NGA and the NRO have been looking at the value of commercial imagery startups for a while. Why does the intelligence community need this office now?

MILLS: Given the revolution in satellite technology, our directors recognized that we needed to come together. We wanted to be able to leverage these new capabilities and new technologies and bring them into an architecture that is not stove-piped. One of the first things that the Commercial GEOINT Activity (CGA) activity will do is come up with a framework of how we can assess these capabilities against our requirements.

MARTIN: We’re pretty excited about this and it’s driven by a recognition that there are a lot of opportunities. But at the same time, as we look to the future, we’re looking at a much more integrated, responsive and flexible architecture. Embedded in that is a significant amount of machine-to-machine processing and queuing. One of our key roles in this joint endeavor is to try to understand what is being offered by the commercial companies and then to see how we can — as seamlessly as possible — integrate it into our future architecture, rather than as an add-on. These companies are centering-up on distributed low-Earth orbit architectures, smaller than what we traditionally had in commercial providers, but with more of them. Their products are very diverse, ranging from just raw imagery — which both we and NGA are very familiar with, in terms of assessing the quality and utility of it — to producing finished products, all the way to economic reports and statistics and that kind of thing.

SpaceNews: Is the CGA a kind of mini-intelligence agency of its own?

MARTIN: Both agencies wanted to be careful that we weren’t requesting a new budget line, new personnel billets and that kind of thing. Everything that’s going into this office is actually a merger of folks — both people and funding — that were already resident in either NGA or NRO. We just felt that given the complexities of this new environment, the technical and acquisition considerations, and the requirements development consideration that we were far better off having those folks merge into a single activity.

SpaceNews: Why does this need to be done jointly and not assigned to either NGA or NRO?

MARTIN: The power is that we’re doing this together, with the best of NGA — understanding the requirements and the intelligence part of the commercial imagery — and the NRO — understanding the acquisitions. We’re much more powerful together than we are separately.

SpaceNews: Is one of the goals of this initiative to be able to incorporate products from these commercial imagery companies directly into your systems?

MUEND: This is an opportunity to better integrate the overarching intelligence collection architecture, by leveraging everything that commercial industry has and minimizing stovepipes.

MILLS: These companies will be able to make sure that we are spending our dollars wisely on capabilities that answer our key intelligence questions.

SpaceNews: Does that mean the NRO may decide not to build a satellite if there’s already a commercial company that’s out there providing the same data?

MILLS: That may ultimately be a decision. If there’s something commercially available that meets our needs, we don’t have to invest in other ways to answer that.

SpaceNews: What have you heard from industry in response to this announcement?

MARTIN: Their reactions are different depending on how they see their business case. Some of them are very focused on commercial business cases and are far forward-leaning; others would welcome major government business. It just represents the diversity of the marketplace out there.

SpaceNews: How does the processing of these images have to change? What does the ground system look like now and ideally what would it look like in a couple of years?

MUEND: The National Reconnaissance Office has built a capability for the National Technical Means architecture — it’s superb. Our architecture today with DigitalGlobe looks good, too, but it’s separate. In the future, the vision is to see an architecture where you could take any of these capabilities from any of our providers, integrate that into the tasking and dissemination of those requirements. A user would come in with one requirement and that requirement would be fed to the best capable collector out there, whether that is something the NRO built or something that the commercial vendor built.

SpaceNews: How can a system like this operate at the speeds you want?

MILLS: Given the amount of capability that we will have, it has to be done machine-to-machine and it has to be done through automation. We need to get that human being out of the loop.

SpaceNews: What does machine-to-machine mean in this case?

MILLS: The machines would be talking to one another, the systems would be talking and the first provider that got to that requirement could satisfy it. There would be discussion amongst the machines.

MARTIN: If you look at some of the new commercial offerors, they are emphasizing persistence over wide areas but at lower resolution with various update rates, for example. Some of them once a day, or twice a week, others multiple times a day. What we’d like to do is machine-process those images to identify changes from the previous image and if those changes are tied to what we think are significant activities then we want to be able to cue a higher resolution system or something that can bring a different type of imagery. You do best-of-breed based on orbit geometry and who has the most timely response. Some of these are temporally sensitive whether it’s a fast developing event or a longer term event.

SpaceNews: What’s the biggest hurdle to this?

MILLS: Today we don’t have the standard interfaces between the stovepipes to allow that to happen. This industry has grown up with a lack of common interfaces.

SpaceNews: A year from now, how would you know if this activity has been a success?

FOSTER: Ultimately, by the degree to which we can actually see evidence that this commercial market sector — transitioning from the current potential that we see — to actually realizing that in a context of creating operational consequence, either in our foundational products or our intelligence products. This is all about really harnessing that potential and realizing it in an optimal way. We have to measure success along those lines. Beyond that, creating more straightforward avenues for emergent commercial market providers to have a point of entry for the intelligence community’s consideration. That’s also a value proposition that we think will translate to industry.