Walter Scott , Founder and Chief Technical Officer, DigitalGlobe

by




Walter Scott was driving home from a paintball game some 20 years ago when he came up with an idea that helped give rise to a $550 million U.S. commercial remote sensing industry that, through applications like Google Earth, has literally changed the way people see the world.

At the time, he was with the U.S. Department of Energy’s Lawrence Livermore National Laboratory, where he worked on space-related projects including the Brilliant Pebbles missile defense initiative. The geopolitical landscape was changing dramatically, with the end of the Cold War and breakup of the Soviet Union imminent. Communications and information technologies, meanwhile, were advancing at breakneck speed.

Scott saw an opportunity to commercialize an activity that until then had been the sole province of governments: the collection of high-resolution imagery by satellite. In 1992, Scott founded WorldView Imaging Corp., which is now known as DigitalGlobe.

With the 1992 Land Remote Sensing Policy Act providing the legal framework, WorldView in 1993 obtained a license to collect and sell satellite imagery sharp enough to distinguish ground objects as small as 3 meters across, detail that was unprecedented outside military and intelligence circles. As Scott describes it, getting the company off the ground proved to be 100 times more difficult than he hoped, but infinitely easier than he feared. Raising money to do something no company had ever done before was the hardest part, he says.

Technology also was an obstacle: WorldView and its competitors were plagued by an early string of launch and satellite failures. Once they got their initial satellites in orbit, the companies struggled financially until the U.S. government stepped in to become a major anchor customer, if only reluctantly.

Today, DigitalGlobe and competitor GeoEye are major providers of satellite imagery to the U.S. government, primarily through contracts with the National Geospatial-Intelligence Agency (NGA) that helped finance some of their satellites. DigitalGlobe operates three spacecraft — Quickbird, WorldView-1 and WorldView-2 — that together can collect more than 500 million square kilometers of imagery a year.

The company went public last May and reported 2009 revenue of $282 million, about 82 percent of which came from government customers. DigitalGlobe recently submitted its bid for the NGA’s  EnhancedView program, which will expand even further the government’s use of commercial satellite imagery.

Scott spoke recently with Space News staff writer Turner Brinton.

 

Did you expect the
U.S.
commercial remote sensing industry would have such a government-dominated customer base?

We didn’t expect that the government would be as aggressive an early adopter of commercial remote sensing as it turned out to be. So when I wrote the original business plan for WorldView Imaging Corp., we actually thought the government would not be one of the first adopters. I think the desire to be able to share imagery with coalition partners was one of the primary drivers for the government becoming an early adopter. In some respects, U.S. government involvement in the industry has been an accelerator of growth, because we’ve been able to profit and plow those profits back into developing new systems and expanded capabilities.

 

Would you say the commercial side of your business hasn’t grown as quickly as you would have thought?

Actually it’s done pretty well. We’ve had double-digit revenue growth over the last five years.

 

Do you think your industry has been hampered by
U.S.
regulations?

Yes and no. I’ll say in general the regulatory regime in the United States has been pretty forward-leaning, except in the sense that it tends to lag behind what foreign capabilities can provide. I don’t think regulations have hampered the growth of the electro-optical imaging industry. So far, we’ve been able to expand the industry dramatically, particularly over the past five years. We’re starting to see international competition emerge that in some cases can offer imagery with resolution that approaches what we can offer domestically. So that’s something we have to pay attention to.

But if you look at radar, for example, I think there’s no domestic commercial radar industry in large measure because of the imposed restrictions. That, combined with the fact that we’ve had millions of years of evolution looking at optical pictures that we haven’t had with looking at radar pictures, and I think basically that’s why you’ve seen radar go abroad.

 

Is there commercial demand for
U.S.
radar satellite capability?

Yes and no. It’s really a function of what the product is. If the product is radar pixels, then it’s probably limited to a small set of government users. But if instead you combine radar with other types of data to do things like finding ships, or using radar as another color, I think there may be value in that. What we’re seeing with WorldView-2 is that the additional four color bands we have really add a lot of information to images. You can pull things out automatically that previously required a human in the loop.

 

Do you think exquisite imagery will always be the domain of the government?

Yes, but the definition of exquisite changes. My definition of the exquisite means anything that is bleeding edge and really pushes the limits of technology. That’s something the government does well. Generally speaking, it has the staying power and the willingness to tolerate risk to push the envelope. In general it’s been the government that’s pushed space technologies along. Commercial companies are really good at operating technologies efficiently once they have been developed and offering more for less. We’ve consistently increased volumes and decreased costs. It doesn’t make sense for the government to do things that can be done commercially.

 

Early on, the companies in this industry sought to get a piece of the billion-dollar aerial photography market. Why do you think that never really happened?

The aerial photography industry is really about custom collection. Satellite constellations historically just haven’t had short enough revisit rates to meet many needs served by aerial photography. But what you’re starting to see now, certainly with our constellation, are daily revisit times. We collected the whole country of Haiti in a single day and actually imaged the country many times over in the period after the recent earthquake there. That’s the kind of currency that begins to enable applications that were previously project-based.

The other piece to the aerial story is you can fly airplanes as low as you want. You can collect 10-centimeter imagery with an airplane if you want to. Technological limitations and government regulations prevent us from doing things like that. So there’s always going to be a tier where aerial is the right answer.

 

You gave away imagery of
Haiti
to relief organizations after the earthquake. Do you have a policy of giving away imagery under certain circumstances?

It’s generally bad business to have a business dependent on the misfortune of others. So we try to do the right thing. We don’t see that as a revenue loss, because we don’t think a lot of relief organizations are in a position to go buy the imagery. Not only are we not losing revenue, but we are doing the right thing, and down the road there will be opportunities aplenty to monetize the imagery we’ve collected over that area.

 

What do you think is the next big thing in commercial remote sensing?

There are really three pieces of the problem. One is getting enough accurate pixels with enough frequency over enough of the globe. I think we are doing a pretty good job as an industry in doing that. The second piece is accessibility. It doesn’t matter how great the satellite is at collecting pixels and streaming them down into an archive if there’s not a good way of widely distributing them in a timely manner. So we’ve been making investments in global content dissemination systems where we pre-stage servers, for example, in Europe, to get data out to users faster. Web services and open standards make it really easy to integrate. The last piece is the things that you do with the imagery. The application of yesterday and today is being able to pull up a satellite imagery map. It’s really cool, I can see my house from space, but what can you now start doing with that imagery? For example, Google Maps doesn’t really pull up elevation information. If I want to do something like put up cellular towers, I need elevation information. So products like that are part of the future. Collecting images is just a starting point.

 

What kinds of technological advancements have been made since WorldView-2 was designed that you might be able to leverage for the EnhancedView program?

I can’t say much because there’s an active procurement going on and we are a public company. I will say when we started designing WorldView-2 in 2003, the main advancement was moving to a larger 1.1-meter aperture and eight spectral bands. There’s still quite a bit of territory to be mined just as a result of those improvements.

 

Have you started doing any work on your next satellite?

I can’t comment.