The Direction of NGA
RETIRED DIRECTOR LOOKS TO THE FUTURE
Lieutenant General James Clapper Jr. retired as director of the National Geospatial-Intelligence Agency (NGA) and from government service on June 13. Military Geospatial Technology participated in a media roundtable that provided reporters an opportunity to ask questions of Clapper about his tenure at NGA and the future of the agency. A transcript of some of those questions follows. Some of the questions have been rearranged from their original order, without altering their content, to provide a logical flow for readers.
RELATIONSHIP WITH NRO
Q: Mr. [Donald M.] Kerr at the National Reconnaissance Organization (NRO) has said that he is switching a substantial portion of his focus to the ground. Given where you have been moving in the past few years, that would seem to mesh pretty well with what you have tried to do. What’s going to be the relationship between NGA and NRO if it is starting to do large quantities of ground processing?
A: Providing that NGA continues to engage, in fact even more pervasively than in the past, with NRO, this is a good thing. There is always the historical and traditional critique that we spend way too much money on collection and not enough money on what happens to this stuff on the ground. Don recognizes that. We recently, after a long period of study, rectified the roles and responsibilities between the two agencies. The original set-up and arrangement between the two goes back to stand-up days in 1996, and so we looked at that and rationalized it so it’s a lot more systemic and there’s a much more finite line drawn now between when the NRO stops and we start.
Assuming that we engage and stay together, and I’m convinced that we will, then this is a good thing. This also emphasizes the efficacy of the importance of something that we are pushing here, which is converge, meaning not to do the ground apparatus— tasking, processing, exploitation and dissemination [TPED]—as a one-off based on an individual collection system. So what we are striving for is called convergence, which is to meld all of this into a single robust sensor-, platform-, and phenomenology-neutral ground architecture. So as new collection capabilities come online, we can ingest that without a lot of traumatic upheaval. “Oh, no, we have to build another unique TPED structure,” which has kind of been the history of this.
This is very compatible with Don’s thrust and very compatible with where we are going. The issue would be marrying up the connect buttons.
Q: Will that require work in the NRO charter?
A: No, I don’t think it would. I don’t think it gets to that sort of cosmic level. The NRO charter has more to do with implementation or execution detail between the two principle mission partners, the National Security Agency [NSA] and us.
Another thing he is trying to do, which is a good thing, is to promote collaboration among the towers. So if I read him correctly, he wants to put a lot of emphasis on multi-INT, on horizontal integration and collaboration, and his charge to NRO is what can we do to facilitate or enable that, which is something that his mission partners have been trying to do as well.
Q: Mr. Kerr’s strategic framework seems to raise the possibility that NRO is going to get much more into data analysis if not the sort of finished work that you do. Do you see this as an issue? Are the two sides going to meld in the coming years?
A: I think we already have a pretty close working relationship between NRO and NGA. The deputy of the IMINT [imagery intelligence] director is an NGA guy. That’s one way we integrate.
I loved Keith Hall’s line when he was director of NRO. The relationship between NRO and NGA now is kind of like the relationship between the taxidermist and the veterinarian. Either way you get your dog back.
In general terms, when the pictures become literal and it’s time for an analyst to look at them, that’s our responsibility in very rough terms. A memorandum of understanding was signed earlier this year.
Q: A recent article in the Wall Street Journal indicated that Lockheed Martin might receive another contract from NRO to build an interim optical satellite. Can you discuss that at all?
A: You should talk to NRO about that. Where we come in, of course, whatever system we buy, we need to be able to ingest the information. Ergo, my speech about the single, converged, robust, platform-centric, phenomenology-neutral as much as possible architecture.
Q: How much have you had to do with the requirements that went into the Federal Enterprise Architecture [FEA]?
A: That’s our thing. We are, as a functional manager, supposed to collect, correlate, synthesize and portray requirements.
Q: There has been discussion of the possibility of a capability gap as a result of the FEA problems. Is this acute?
A: Whenever you have the possibility of an interlocking series of disasters, we have always had the prospect of a, quote, gap. The possibility frankly is remote.
I would say that’s another argument though for having as holistic an approach as we possibly can. I view, for example, commercial imagery as somewhat of an insurance policy to mitigate against potential gaps. That’s why it’s an important part of our architecture.
RELATIONSHIP WITH DHS
Q: Could you address the relationship between NGA and the Department of Homeland Security [DHS] and how that has evolved in the past several years?
A: In NGA’s case, all of the classical and historical products, services and solutions that NGA, the National Imagery and Mapping Agency [NIMA] and predecessor organizations have long done in an overseas, foreign environment, principally in support of the military, those same tactics, techniques, processes, solutions and applications can and have been overlaid in a domestic context. They all kind of fit. GEOINT probably lends itself more than any other of the collection disciplines to that sort of thing. So it’s been a natural. The historical heritage for that is in disaster relief. For many years, NGA and predecessor organizations have assisted and support the aftermath of earthquakes and fires. This sort of capability morphed after 9/11, and with the stand-up of the Department of Homeland Security, has become a routine thing.
The epitome of this occurred in the aftermath of Rita and Katrina. That was a huge thing for us. We had a lot of people deployed in the affected area. What we found was just as we had provided that common operating picture for the military overseas, we did the same thing here. We did a lot of before-and-after work, which helped facilitate response, rescue, recovery and restoration of the infrastructure and all of that sort of thing. It all starts with that geospatial depiction of what the situation is.
We have a team now, one of our NGA support teams, that is embedded in the Department of Homeland Security, as we do with all of the commands and services. It’s there as our forward footprint but then they can reach back into the larger NGA.
Q: Is there a formal consulting relationship now that DHS is getting more and more into unmanned aerial vehicles and surveillance technology? Do they turn to you?
A: We have had a lot of dialogue with them about that, particularly with respect to how you task, process, exploit and disseminate. We have had interaction with DHS on those things.
Q: Are there better procedures now for working with DHS?
A: Yes, there are. We must have a lead federal agency when we are going to bring to bear intelligence resources in a domestic context. We have pretty much refined the process for domestic imagery requests. We have a team embedded over there, so we are in tune to them. They are embedded in components of DHS, so when I say DHS I don’t mean just headquarters. We have people in the Secret Service, Coast Guard and FEMA. We have people embedded and they liaison with us, which is invaluable because then we know what is coming up.
All special security events that we have in the country—the Super Bowl, the World Series, political conventions—we are normally called upon to support that and we normally deploy a team. If it’s of sufficient magnitude, they will provide that common operating picture.
Q: How do you see the agency’s role when it comes to state-sponsored or individual proliferators of weapons of mass destruction?
A: We are an enabler. This is a very difficult problem. We are important but one source of the data and the information that it takes to arrive at some of the difficult determinations about any proliferator’s weapons capability. Our role is the scene-setter and deriving from the mission of geospatial intelligence what it is that we can glean from that, but we are certainly not the endall answer.
RELATIONSHIP WITH STRATCOM
Q: Have you talked with Gen. James Cartwright [commander of U.S. Strategic Command (STRATCOM)] about Global Strike plans that would require that sort of technology?
A: We talk frequently. He tells us what to do and we say, “yes, sir!” He’s very much a proponent. As he stood up his functional component commands, most notably for us the ISR component command, we have people embedded in that. What this represents is a great potential here for better and more efficient use of the totality of resources available.
He’s also a big proponent of standards. He’s made a huge push for that, which is a good thing. In my separate hat as the functional manager for the National System for Geospatial Intelligence, that’s one of the big items we need to promulgate and enforce standards.
Q: What needs to happen before the Prompt Global Strike can become reality?
A: That’s not really my department. My main interface with STRATCOM is on a number of fronts, but principally on ISR, not Global Strike. We have a team embedded at STRATCOM headquarters, so the extent to which they require targeting support, access to our digital position database, that sort of thing, then we can facilitate that.
DIRECTOR OF NATIONAL INTELLIGENCE
Q: There’s been a lot of chatter on the Hill and other places about whether the director of national intelligence [DNI] is bringing people into line.
A: We certainly need the supervision.
Q: Is the DNI getting the job done?
A: In fairness, people had a very high expectation of what was going to happen here when Congress passed the intelligence reform legislation. The way I think about it is where were we in 1948, a year after the enactment of the National Security Act of 1947? Well, we weren’t where we are today in terms of the evolution of the Department of Defense and military defense. A lot has happened in the 60 years almost since 1947. The DNI officially stood up over a year ago. If you put it in that context, they have done a lot.
A big thing for this agency, in my view, was deciding on this long-running gun battle on what’s geospatial intelligence versus MASINT [measurement and signature intelligence]. Well, that was huge for us and they made a decision on it. There has been a sorting out of roles and responsibilities that have accrued by virtue of the changes with CIA. So certain authorities and responsibilities that continued to be conducted by CIA after the stand-up of NIMA in 1996 now have been vested in this agency.
So from my standpoint, they have done some good things. But instant miracles? Curing world hunger in a year? I don't think so. Those are unreasonable expectations.
Q: Critics who are boosters of the CIA have criticized removing the analysis of geospatial imagery from the CIA as far back as the establishment of NIMA in 1996. Can you share your thoughts when it comes to defining responsibilities between NGA and CIA?
A: When I danced off the stage in 1995 as the director of the Defense Intelligence Agency [DIA], I was an abject opponent of the absurd idea of having an imagery agency. What a dumb idea that is. Of course, as a classic irony, what goes around comes around, and I have become a firm believer. The term "geospatial intelligence" wasn’t coined then.
Well, a lot of things happened and a lot of studies were done on that. A milestone event was the NIMA commission and the NIMA commission report. NIMA had the distinction of being the most heavily studied component of the intelligence community in the first five years of its existence. By one man’s count, some ten different studies were done on NIMA and how it was doing. The charge I was given in the summer of 2001 was to implement the NIMA commission report.
The whole objective was the melding and synthesis of mapping and charting on one hand and imagery analysis and imagery intelligence on the other. The logical [outcome] was to put those resources together. That meant to take the National Photographic Interpretation Center, which was a component of CIA, as well as the Office of Imagery Analysis within CIA, and to make that into a national agency that would have the institutional clout and institutional resources and institutional advocacy parallel to the National Security Agency. That was the whole debate and discussion that went on after Desert Storm in the mid-1990s.
I have become a zealot for this. I really believe it is the right thing to do. The issue then becomes one of objectivity and preserving and nurturing that tradecraft, which is another reason why I think this is a good thing. A lot of people use the term stovepipe pejoratively. I think stovepipes have merit in that they are the home of the tradecraft. There are certain skill sets that are required to perform geospatial intelligence. There are certain skill sets that are required to perform signal intelligence. The agencies emerge as the harbor for the development, the nurturing and the tradecraft that it takes to conduct these disciplines. In that respect, I think there is strength in that.
When I was DIA director, imagery analysis was kind of second-class and always taking a back seat to all-sources analysis, which is understandable because that’s their mission. Within the GEOINT stovepipe, and I use the term not pejoratively, that is the central core of our mission and then we are responsible for training people and advancing them in career development and advancing the tradecraft. We get much better focus that way than if it is a secondary enabler for somebody else. It has a tendency to get neglected somewhat in that context. That’s the way it was when I was director of the DIA.
Q: What do you think of the nomination of General Michael Hayden as director of the CIA?
A: He’s a great guy. He and I are personal and professional colleagues and have been for many years. I think he will impress all of them.
Q: There has been a tremendous amount of talk about infighting between CIA and the military intelligence agencies over the past two years on the Hill. Are the folks on the Hill seeing fights where there are not?
A: I don’t see that. I think in some ways there have been some very creative and unprecedented things done where the military and the CIA collaborate in the field today. I was chief of Air Force Intelligence in Desert Storm and for the next four years or so the director of DIA, and one of the major critiques that emerged from General [H. Norman] Schwarzkopf personally was his complaint about the lack of coordination between U.S. Central Command and CIA.
One of the fixes instituted, which still exists today, by the then-Director of Central Intelligence Bob Gates, was setting up an Office of Military Support, headed by a flag officer, inside CIA. I had a lot to do with picking the first guy that was in that job. That has been sustained now for more than 15 years. They have always had a flag officer from the military in either intelligence or special ops backgrounds in that position. It has done a lot to facilitate collaboration between the agency and the Department of Defense [DoD].
PROGRAMS
Q: Could you give us an update on the status of the GeoScout program and how it fits into the focus on analysis?
A: It’s actually going very well. We are finishing off Block One, and we are into Block Two, which is on-schedule and at-cost. The major focus of Block One was the infrastructure— the building blocks or foundation if you will—given what the agency inherited in 1996 in the way of infrastructure when it stood up, which was bad.
We decided that starting with that was the most important thing, and then Block Two was information management, focusing on how to manage the ingestion of the increasing volume, variety, velocity, veracity of data that we are going to have to deal with here in the next decade.
Our focus then is to finish Block Two, but then after we are going to be doing what we have been doing all along anyway, which is to acquire specific discrete increments of capability. I think that we are just going to formalize that process and not really have any more blocks but just do incremental improvements, which I think is much more in tune with what industry would do, given an IT context where the technology turns over so rapidly. So hopefully we can be a bit more agile.
Q: Do you envision indefinite-delivery, indefinite-quantity [IDIQ] contracts?
A: I don’t know what the contract vehicle would be. We will continue with our same approach contractually, but the business of having big blocks like we are buying F-22s or something, which we are not, doesn’t really lend itself to a more agile IT acquisition approach.
Given that we have these major capabilities— finish the infrastructure and information management—and then as we get into applications and tools, which are going to be more specific discrete capabilities and not so monolithic. Whether for the whole enterprise or for a discrete set of analysts, we are sort of getting to that stage.
Q: When is the information management portion supposed to wrap up?
A: I think it’s 2009.
Q: Regarding architecture, does FIRES fit into that or not?
A: Well, it really doesn’t. I think the spirit of FIRES probably lives on. There is a DNI study going on—in fact a number of studies going on in which we are all participating— that would get to an objective architecture that would be more unified. So the FIRES program per se is a programmatic casualty of last year. I think the spirit and the intent of FIRES lives on.
Something like that has to be a community thing. It’s not a bill that any one agency could or should foot.
Q: You have mentioned the possibility of expanding the NextView program. Could you tell us what that might involve?
A: By expanding it, I think the notion was perhaps follow-ons. That is kind of way out there again given the vagaries of our funding system and the way money is appropriated. I am anticipating, being an optimist and a huge believer in the value of commercial imagery, that we will continue to sustain a pretty healthy engagement with the commercial imagery industry.
So delays are understandable. Those things happen in programs like that. The companies have a good handle on what has to be done to fix their issues.
TECHNOLOGY
Q: How do you see the agency moving forward with the speed of data and persistent surveillance, things that have been on the forefront of the agency’s concerns?
A: A major concern for us at the agency coincident with getting ready for Operation Iraqi Freedom was for the first time ever bringing in live feeds from unmanned aerial vehicles [UAVs]. The Predator and Global- Hawk have now become a regular thing. That of course is the harbinger of persistent surveillance. The advantage here is bringing to bear the expertise that resides in this agency, performing in our combat support agency role and supporting those forward deployed in the commands, helping with target characterization, identification and whatever.
So we are directly participating in that effort. That’s one of a number of new things in NGA. Others being the proliferation and importance of commercial imagery and what we call advanced geospatial intelligence.
Q: You’ve mentioned incremental funding for specific programs in the past. Is that something to look at as you move forward for most of your acquisitions?
A: Well, we get money appropriated once a year. So we have to acknowledge that. The Congress giveth and taketh away. So the most sensible thing to do to me is to build an acquisition strategy that is congruent with how you get money. They don’t fund us ten years ahead. Industry would love it if they did it that way, but they just don’t do it that way. That’s why I’m a believer in smaller, definable, more discreet acquisitions. We are sort of getting to that stage in maturity with GeoScout and our modernization here where we have reached that point where if we ever get any more money for modernization, we have got something to show for it.
Q: With that in mind, what do you think are the most important technological developments to move forward on?
A: Automatic target recognition, automatic feature extraction, data compression—just to name three of them.
We are going to have a huge volume of data to handle. If you look ahead at the next ten years, there are huge increases in data, the number of platforms, the number of sensors, the phenomenologies that we are going to have to take care of, and the rate at which we are going to have to ingest it. There is no way we have enough eyeballs right now to look at all of that. So to the extent that you can automate even a 75-percent or 80-percent solution with automatic target recognition and automatic feature extraction.
Any major weapons system in DoD today requires an infusion of accurate geospatial intelligence. We cannot manipulate it by hand. We must have automation tools that will take some of the burden off of analysts, so that they can focus their time and attention on things that only the human brain can deal with.
The challenge of course is that adversaries can learn. So appreciably for us, denial, deception, camouflage and concealment are a challenge, which is why it is important that we be more holistic in terms of the numbers and types of platforms and phenomenologies that we exploit, so that we torture the pixels in all forms.
Q: There has been some chatter about how important are pictures from above, given some of the things that we have been able to do. Can you discuss that?
A: Just with the airplanes versus satellites issue, if you are not concerned about flying airplanes over denied areas, then you don’t need satellites. Well, I don’t foresee that happening in the near future, that we are not going to need capability that is systematic and global and can reach any point on the face of the earth, whether it is denied or not.
Q: Can you elaborate more on the relationship between satellites and unmanned aerial vehicles?
A: The fact of the matter is that in general terms there are complementary attributes with satellites and airplanes. Satellites stare. Unfortunately, we can’t do the Tom Clancy thing. I wish we could. But you have the advantage of global coverage. You are able to reach any point on earth, within reason, within 24 hours, and you have access to denied areas.
With airplanes on the other hand, particularly unmanned ones, [they] give you the advantage of being able to stare for extended periods persistent but for a comparatively smaller area. The trick is how do you meld these complementary attributes of both platforms so that you are not dependent on one or the other. You want the mix.
The same is true of commercial imagery. Commercial imagery has some distinct advantages, one of the main ones being that it is unclassified. You can share it in a coalition context. You can share it for public diplomacy and you can use it in a homeland security context. And we did this for Rita and Katrina. We put it out on the Web for the public to look at. Again, it doesn’t have quite the agility and responsiveness and the sensors aren’t as capable, and that’s kind of deliberate, but it is a very valuable complement to our system.
If you think about this in terms of complementary attributes, the mix of sources and the mix of platforms are important.
Q: Whose UAVs are you using now?
A: We don’t use any. We don’t task any UAVs. We provide the eyeballs and the analysts that look at it. So it’s Predators and GlobalHawks from the services mostly.
Q: As the GIG bandwidth expansion is going forward, do you have enough bandwidth now?
A: No, you can always use more bandwidth. That’s like asking me if I have enough money now. We would always like more. Contributions are always accepted.
When I first got into this business 40-plus years ago, I can think about how difficult it was to move data around now. The notion of moving pictures around electronically was not possible. We did not have satellites during my first tour in Vietnam as a young intelligence officer. Automation in those days was an acetate grease pencil and two corporals. That’s how we automated. I remember sitting on one end of a terminal that didn’t work whenever there was rain and lightning.
Fast-forward to Desert Storm, and I was chief of Air Force Intelligence then, we had come miles since then. I distinctly remember the frustration over the nine separate secondary imagery transmission systems. That was the way we moved imagery. Very slowly and inefficiently. What we can move around today is unbelievable.
But would we like more bandwidth? Would we like huge bandwidth down to the platoon commander in the foxhole? Sure, but we don’t quite have that yet. But it’s vastly improved over what it used to be. The GIG Bandwidth Expansion is hugely important to NGA. We are dependent on it for our future architecture. As we field GeoScout, we are profoundly dependent on the GIG Bandwidth Expansion.
Q: Is there any issue with the balance between the data you are getting and that you are actually able to use effectively and manipulate?
A: The objective there frankly is to have this huge reservoir, archive or library or whatever you want to call it, where the data can be there and extracted when you need it. That’s kind of the way that we do business today. We make a huge investment in having our own people out there at the pointy end of the stick. We are not unique in that as all of the combat support agencies are doing this. Those people understand what is available in that reservoir and can extract it or have the people tending the reservoir back home help build and tailor products, services and solutions for people at the pointy end of the stick.
The main thing is to populate the data or the layers, as we say particularly that foundation layer of the earth, whether terrestrial or maritime, over which other forms of information or intelligence can be overlaid.
Q: This platform-independent convergence strategy that you mentioned, how do you see this playing out, for example, in acquisition? What new strategy should NGA or DoD take?
A: The strategy that we are taking with GeoScout is to promote a converged architecture, so it’s not tied to a specific platform.
Q: Does that require going back and making your existing data backward compatible?
A: I think your question deals with the whole issue of meta-data tagging. I don’t think it’s affordable or feasible to go and relabel all of the data that we have archived. You are talking about millions and millions of images. We can get to it, and we have done a lot just to automate the process by which we can dredge up older imagery. Clearly, that is part of the process here, if you can compare what did this spot on the ground look like yesterday, last week, last month, last year, or five years ago. But I don’t think we are going to go back. We are doing what is current.
Q: What do you see as the agency’s biggest challenge in the next six to twelve months?
A: The biggest challenge for us is that the eastern part of the agency has fallen under BRAC [Base Closure and Realignment Commission]. We are going to move to a consolidated and modernized campus at Fort Belvoir. The law mandates that we will be in by October 2011. That is not very far off. This is a huge thing for the agency. There will be all kinds of decisions that we have to make about systems. What do we retire in way of legacy or heritage?
This is going to have a huge impact on business practices. Since we can design a facility that is optimized for the mission instead of just a collection of buildings, an ad hoc cats-and-dogs set of leased and government facilities. Some of them are okay and some of them are old and falling apart. They are not designed for our mission.
Having two-thirds of the agency in the east and one-third in Missouri is a challenge. But that, to me, is going to be the major issue my successor is going to have to work. It isn’t just brick and mortar. It isn’t just a new campus and a new facility. It is going to have a profound impact on the conduct of the geospatial intelligence mission. ♦
