Scott Large


National Reconnaissance Office (NRO)

The acquisition reform movement of the 1990s was supposed to reduce the cost of national security space programs via commercial development practices, but wound up doing the opposite. For the U.S. National Reconnaissance Office (NRO) the result was damage not only to its reputation for technical wizardry but also to its ability to carry out its mission.

Scott Large holds up the Future Imagery Architecture (FIA) optical and radar imaging satellite program as an example of what happens when too much systems-engineering and program-management responsibility is shifted to prime contractors. The NRO awarded Boeing the contract to build the FIA satellites in 1999, only to cancel the optical portion of the program in 2005 amid severe technical problems leading to lengthy delays and cost growth in the billions of dollars.

By that time the NRO had lost much of its skilled work force, accountability was scattered and the internal processes that brought the organization success in the past had deteriorated, Large said. In 2003, the agency began instituting organizational changes aimed at getting its swagger back.

Historically the NRO was divided vertically into three separate mission directorates – imagery, signals intelligence and communications – each responsible not only for building its own satellites but also for the associated ground systems and operations. Now the agency is organized more along functional lines, with separate directorates for ground systems and operations that consolidate the three mission areas. Large considered doing the same for satellite development, but recently decided to maintain three separate organizations for that activity.

The agency also created two chief operating officer positions responsible for ensuring that processes are followed.

“That has put accountability back into the process, which is absolutely crucial,” Large said.

In terms of future systems development, much remains up in the air. The NRO is in the early conceptual stages of an FIA follow-on system, and its radar strategy is unclear following the cancellation of the Space Radar, a joint program with the U.S. Air Force. Meanwhile, Congress recently derailed NRO plans to procure two commercial-like imaging satellites under a program called Broad Area Space-based Imagery Collector (BASIC), whose capabilities some believe are best provided by the private sector.

Large, who took the NRO’s top job in October 2007 after serving as director of the Source Operations and Management Directorate at the National Geospatial-Intelligence Agency, spoke recently with Space News deputy editor Warren Ferster and staff writer Turner Brinton.

The NRO has made significant organizational changes. What is the next step?

Roughly 60 percent of my work force has been here less than five years, so we don’t have as much graybeard experience as we’d like to. I will put the engineers, system engineers and program managers that we have at the NRO up against anyone technically. These kids coming out of school are incredibly smart. In order for them to use the smarts they have, they’ve got to build that scar tissue. I’ve got to find a way for them to get those skills and experience.

Why does the nation need BASIC?

In terms of where we are with our constellation now, it is the next step in the architecture that the director of national intelligence and the secretary of defense have agreed to move forward with. Commercial systems today augment the national systems, pretty much in volume, especially for non-time-dominant activities. What the Defense Department is asking us to do with BASIC is do a better job of integrating a commercial-like platform into the government architecture so other government platforms can do more specific activities that the commercial platforms can’t.

There are several aspects to the decision: access to the data that gets collected, content of that information and timeliness in getting it to the user. It’s one thing when you’re talking about map production. When you’re talking about imagery that needs to be analyzed in real-time for things like targeting in mission operations, there are timeline differences. BASIC will be inserted into the national architecture to provide for the operational needs that require access to the data and timely delivery – faster than you’re going to get today in the commercial architecture.

Why can’t that be done with commercial satellites?

The analysis that was done concluded the best way to approach this would be to integrate two government-owned platforms into the existing architecture.

What’s the timeframe for the follow-on to FIA?

We’re going to have to understand what we need to build and how we want to approach building it, probably beginning in the next three to four years. We have to be solidly under way with preacquisition, which is a key topic here. With FIA, we did not do adequate preacquisition for a number of reasons, including budget and schedule pressure. The challenge we faced when we did that was not driving down technical risk adequately before we got started. You need to spend a year to two years up front before a decision can be made to commit dollars to go buy something. I would like to think we could be at a point to make that decision to buy three years from now, which means I want to start preacquisition in the next year or two. But the budget profile will determine the timeframe of any capability.

I think the nation needs to come to grips with how much we are going to commit to overhead space reconnaissance and how much we are going to commit to other things like airborne reconnaissance. This is a very critical point the intelligence community and Defense Department need to address because that commitment means time and dollars, and this is not a cheap business. I think we’ve proven that attempts and experiments to try and do this faster, better and cheaper can get us in a lot of trouble if we don’t pay attention. This will never be cheap. It’s a matter of how much do we want to invest to do this?

Might the FIA follow-on contracts be awarded on a sole-source basis?

In my view, we’d look at any contracting acquisition approach. I can’t tell you today that it would be competitive or sole-sourced. We have a lot of oversight we have to go through before making any decision like that with the director of national intelligence and the undersecretary of defense for acquisition, technology and logistics. Decisions were made when we started FIA in 1997 to take a different approach for this, which was the acquisition reform approach. There was a lot of pressure to do something different from the presidential administration, Congress and within the NRO.

This next generation of systems, whatever it may be, is yet to be defined. I believe the architecture for the future has to take into account our national security space industrial base, which is suffering right now. We’re not a volume customer. We must establish within the industrial base a level of business that justifies those companies wanting to stay in the business.

Why was Space Radar not a workable program?

I think the program was too large when it first came out of the gate. The end-state objectives, the budget it would have required and the delivery time to get there were not tenable. Trying to be all things to all people may have been part of it.

Will we ever see a joint Defense Department-NRO program for radar imagery collection?

I certainly hope so. We’re going to go in a direction based on what requirements we’re asked to meet, whether it’s from the intelligence community or whether there are requirements from the Defense Department beyond what commercial systems can satisfy. Commercial systems may free up dollars for the NRO, working with the Air Force, to figure out how we could satisfy that next tier of requirements.

What’s your philosophy on the question of whether we should build less complex satellites?

It’s very mission specific, mission dependent. There are advantages in going with smaller platforms in terms of complexity and design life. When we make decisions for the future, it’s got to be based not only on requirements and performance; it’s got to be based on the larger picture of whether we can sustain industrial base production lines and drive down cost. In some cases that will drive you to a smaller platform. If the requirements are out there for an exquisite capability, the physics will drive you to a size of spacecraft just to meet the requirement of the data they want to collect. We go back and forth on this, and frankly, the philosophy of BASIC is a smaller, simpler, cheaper platform that we can produce probably more rapidly and leave the challenging, exquisite types of requirements to a different type of platform.