John Olesak

Vice President of Geospatial Intelligence, Northrop Grumman Information Technology Intelligence Group


he last decade has

brought tremendous

change in the way

geospatial intelligence products are delivered and consumed.

Technology has advanced at such a rate that the bottleneck for getting this

information to military forces

is no longer at the collection point, but in the

number of systems and analysts that can


and distribute

data at any given time.

The emphasis

increasingly is on systems able to handle imagery regardless of its collection platform as well as combine geospatial with other types of information such as electronic intelligence. Meanwhile, the geospatial-data user list is growing: The U.S. Department of Homeland Security (DHS), for example, is creating a National Applications Office that will handle domestic civilian requests for classified satellite imagery.

These developments are driving changes at both the U.S. National Geospatial-Intelligence Agency (NGA), which is still in the midst of an infrastructure overhaul, and the National Reconnaissance Office

, which is reorganizing and placing more emphasis on its ground systems.

John Olesak says his company, which provides geospatial solutions to U.S. government customers, must stay agile to respond to this dynamic environment. He spoke recently with Space News staff writer Turner Brinton.

In the area of geospatial intelligence, who is Northrop Grumman’s biggest customer?

The National Geospatial-Intelligence Agency

is Northrop Grumman’s biggest customer for geospatial intelligence. But within the Intelligence Group, the NGA is second to the National Reconnaissance Office


Has the National Reconnaissance Office’s

new emphasis on ground systems had significant implications for your company?

Yes it has. We’re very close to those types of changes in the customer community, whether they are from the intelligence community or the Department of Defense. We know that things will not remain static as outside influences change and technologies are available to support new ways of doing things. So we’ve been very close to the movement toward more of a ground-oriented architecture, and we have adapted and adjusted our programs and our support to stay consistent with that. In many cases, we can offer suggestions as industry partners on how that type of change in the landscape can be better accommodated.

How are commercial applications like Google Earth changing the way people use geospatial intelligence?

It’s making it a lot more prevalent and user friendly, and it’s making it a lot more visual. When we saw Google emerge a couple of years ago, it gave us a tremendous visualization capability that we could download onto desktops. From an analyst perspective, we saw that we could immediately adapt to that commercially available visualization capability. But there are a lot of things that


won’t do to support some of our more traditional military customers. So what we’ve been doing is adapting Google visualization technology to military and intelligence community needs and deploying that visualization technology out onto our networks for our analysts as well.

Is Google Earth a revolution?

I don’t think it’s a revolution, as there have been research programs developed over the years that were moving in the Google direction. I think the revolutionary part is capturing that and making it available to individuals and the general public as a visualization technology and promoting the research and development without having to subscribe to Google. Getting it for free and downloading it to your desktop stimulates our appetite for geospatial intelligence in the private sector.

How can the military take advantage of Google Earth?

Somehow we have to come up with ways to use imagery of higher and higher quality over larger and larger land mass areas and then look for ways we can update that imagery.

If you think about a military application, an image that is two years old isn’t really much good for today’s mission. So we have to take advantage of things that we have developed over the years like image libraries that are constantly being refreshed for intelligence and military applications and take the currency of that imagery and move it into a Google-type application.

There are also things behind that imagery, commonly referred to as metadata, that tell you things about what you are looking at. You see a little of that in Google, where it will give you points of interest or tell you something about a feature. We have a need for much more of that

to support military and intelligence applications.

How large a customer for your services is the DHS?

Not as large as some of our more traditional intelligence community customers, but it is still a significant customer base. As DHS continues to mature and expand, we’re continuing to offer solutions.

Do you expect

your DHS business to grow

once the National Applications Office is up and operating?

I think so, absolutely. I think as DHS stands up different offices and adapts to the integration

of many different government organizations, we will continue to see enterprise applications and things that we have done in our traditional intelligence and defense areas being adapted to first-responder solutions for DHS. Things like disaster relief, disaster preparedness and 9-1-1 communications are very transferable into our homeland security environment.

U.S. Marine Corps Gen. James Cartwright, vice chairman of the Joint Chiefs of Staff,

has said

the U.S.

intelligence architecture

lacks integration and collaboration because industry has not been given the proper incentives

. Does the way

these systems are acquired need to


, and is it realistic to think that

can happen


If you look at what NGA is responsible for as the functional manager for geospatial intelligence, they have a concept called the National System for Geospatial Intelligence. It isn’t a point solution, but an integration of capabilities from a variety of different contributing programs that make up the larger system for national geospatial intelligence. We don’t incentivize program managers for the success of that larger system. If we’re going to get to

an interoperable state, we really need to think about the value of the National System for Geospatial Intelligence.


we get into joint force warfare and coalition warfare we continue to need to scale up in terms of where we put the focus and where we put the incentives. So I think that need is recognized and there is a movement in that direction. We’re already seeing a little of that in some areas where there are large systems being developed in the services – we just need to scale up in some

cases to get that interoperability on larger enterprise levels.

What kinds of geospatial intelligence systems does the United States need more of?

I think we’re going to need to take advantage of imagery from a variety of different sources. The notion of always needing bigger and better satellites is probably not correct; it doesn’t lend itself to providing

imagery information that is usable by everybody. And some of the larger systems require incredibly long lead times to develop and launch and get into operation. Smaller satellites that are more agile and quicker to develop can still produce high-resolution imagery over a broad area. I continue to see it as a mix of capabilities that will allow us to adapt to changing mission requirements.

What is the biggest challenge facing the geospatial intelligence industry today?

It’s somewhere between our ability to collect and our ability to exploit. I think we’ve seen over the

last several years that our collection capability has far outgrown our ability to exploit and analyze. So I think part of our issue is going to be closing that gap. We can collect a lot, but how much of that data is truly meaningful, and how much of that data can we get into the hands of the right analyst at the right time so that he or she can turn that collected information into intelligence that supports a decision?

Is it partly a matter of being more efficient in data collection?

Yes. There

are still things we can do in the automated target recognition environment that would take some of that out of the hands out of humans and allow machine techniques to do some of that early filtering. That cuts down on the volume of data that ultimately finds its way onto the workstation of an analyst. Better image compression, search capabilities and archiving capabilities are also high priorities.

Will these capabilities ever be good enough to make human analysis unnecessary?

No, there will always be a human element in that decision-making process, but you do want to reduce the amount of human time in that chain and develop better and better technologies you can rely on.