300 N. Washington St.
Suite B-100
Alexandria, VA 22314
info@globalsecurity.org

GlobalSecurity.org In the News




NewsFactor Network June 27, 2006

Does Google Earth Reveal Military Secrets?

By Barry Levine

"I think this is about the end of the road for commercial [satellite-imaging] systems," says John Pike at GlobalSecurity.org. At higher resolutions or greater update frequencies, he says, "They would begin to replicate military systems, and the government has no intention of allowing that."

For centuries, the highest vantage point in a battle provided information that could pierce "the fog of war" and possibly tip the outcome.

For about a year, the satellite and aerial images in Google Earth and Google Maps have been informing and entertaining civilians across the planet, setting the stage for competing services such as Microsoft's Virtual Earth and Yahoo Maps.

The technology that powers these free programs has become sophisticated enough to render terrestrial details in astounding clarity, in some regions at a resolution of 0.6 meters per pixel, taken from nearly 300 miles up.

The growing popularity of such applications, however, has led to questions about whether they are, in fact, compromising military secrets and national security.

Officials in India, South Korea, Australia, Thailand, the United Kingdom, the Netherlands, and Israel, among other countries, apparently think so. Each nation has expressed alarm that these "interfaces to the planet," to adopt Google Earth's self-billing, could possibly provide no end of useful information to terrorists or hostile governments.

An Eye Toward Danger?

In 2005, for instance, Indian President A.P.J. Abdul Kalam showed satellite and aerial photos of sensitive locations in his country -- pulled from the Internet -- at a conference of that country's police officers. He warned about the possible dangers that Google Earth and other "open-source intelligence" pose, especially for developing nations.

India has been in an on-again, off-again state of near-war for years with its neighbor, Pakistan. It has also been a target of major terrorist attacks.

"If you talk to the Indian government, they think [commercial satellite imaging] is the worst thing that ever happened," says Theresa Hitchens, director of the Center for Defense Information in Washington, D.C. "They think it can help their enemies locate facilities."

In Sydney, Australia, the operators of the Lucas Heights nuclear reactor complained last year that Google Earth increased the likelihood of a terrorist attack on their facility. In the UK, officials in charge of nuclear plants have tried to block the availability of satellite photos of their facilities.

South Korean officials, representing another country in long-term near-war, have expressed concern that satellite photos of military installations and the presidential Blue House could offer informational aid to North Korea. The photos, they said, might even violate South Korean security laws.

But, like officials in other countries worried about satellite imaging, the South Koreans have acknowledged their inability to enforce restrictions against foreign companies.

No Going Back

Some experts have argued that the anxiety is somewhat overblown. According to John Pike, director of GlobalSecurity.org, a military and intelligence information Web site, the overhead imagery that's available online doesn't provide information that cannot already be easily obtained in an open society -- from other image providers, from governments, or from other sources of public information.

"The cat's already out of the bag," Pike says. He adds that the kinds of images found on Google Earth are "not particularly useful" from a military perspective, and that there is "a negligible risk" to a free country's national security.

These twin arguments -- that photos can be obtained from other sources and that, regardless, they're not particularly useful to evildoers -- are the main defenses against charges that satellite-imaging applications pose security risks.

Satellite imagery can be purchased from any number of sources. Targets can be reconnoitered by land or flyovers, and tons of related public information, such as maps, are readily available.

"Anyone who flies above or drives by a piece of property can obtain the same information," a Google spokesperson said in April when responding to the concerns of the Indian government.

In an open society, trying to restrict sources such as Google Earth may be a fool's errand, Pike says. However, if a country is tightly controlled -- such as North Korea, which keeps a close watch on the information that enters the public domain -- then easy Internet access to satellite images constitutes a crack in the armor.

In fact, as Mark Brender, vice president of communications and marketing at the satellite-imagery provider GeoEye, was being interviewed for this article, he was interrupted by another call. It was a request from CNN to obtain satellite photos of the area in North Korea where a long-range missile was reportedly being fueled.

A Practical View

If the same or comparable imagery provided by Google Earth can be found elsewhere, it raises the question of whether Google Earth or similar services are of any practical use to organizations such as al Qaeda.

Some people, like John Pike, say the information is not critical. "The evildoers did not need satellite imagery," he points out, "to find the World Trade Center." But CDI's Hitchens says that such imagery, even if not unique, could still help a terrorist group. "You can combine satellite images with information on the ground and get useful information," she says.

The extent of that usefulness might be compromised, however, by the age of the photos. Commercial satellite imagery on the Internet is not displayed in real-time.

"We have a 24-hour hold on all images," says Chuck Herring, director of corporate communications at DigitalGlobe, a major satellite image provider for Google. He adds that, in practice, most of the images his company supplies to Google are at least months old, which is typical of image providers.

Some of the imagery on Google Earth or Google Maps is actually several years old. For any time-dependent information, such as troop movements, the best these images could do would be to serve as a record of recent history, not a real-time window. Such outdated images, says CDI's Hitchens, would be a bad source of information for any military campaign.

"In 1999, during the Bosnian War, we bombed the Chinese Embassy by mistake because of outdated maps and aerial images," she says.

For the operators of the Australian nuclear plant at Lucas Heights, which had originally requested that the Google Earth images of their facility be censored, the photos' age -- plus their availability from other sources -- apparently lessened the risk posed by putting them online. The operators eventually concluded that, because the images of their facility were two years old, they did not reflect the current infrastructure and so were not a security breach.

Shutter Control

Satellite-imagery providers follow no clear policies on when and if they should censor themselves. DigitalGlobe says it does not censor images. "If it's a military secret, we don't know about it," Herring says.

However, when officials at DigitalGlobe know they are dealing with sensitive areas, according to Herring, such as Iraq and Afghanistan, then the company tries to monitor the image-taking. He declined to discuss exactly what measures DigitalGlobe takes to avoid revealing such things as troop positions.

And if the U.S. government wanted to censor the images, it has at least contractual authority.

"Under the terms of our license with NOAA," says GeoEye's Brender, referring to the National Oceanic and Atmospheric Administration, "we can image anywhere in the world. And there is a provision that the U.S. government can interrupt commercial service." This protection, which involves closing the satellite's eye when it is tasked over certain areas, is sometimes called "shutter control."

But the U.S. government has never invoked shutter control against either GeoEye, which supplies Yahoo Maps (now in beta) and Microsoft Virtual Earth, or DigitalGlobe, a major provider to Google Earth. Instead, in at least one case, Uncle Sam took what might be called a free-market approach.

When the war in Afghanistan was starting in fall 2001, Brender says, the Pentagon bought up three months of exclusive rights to all of GeoEye's imagery of Afghanistan. In January 2002, when the government's exclusive licenses were up, all of those images were made available commercially.

Google, whose satellite images are perhaps the best known on the Net, has a no-modification policy. "Google doesn't doctor or change the imagery that we receive from our third-party satellite data providers," the company said in a statement. "If users are seeing 'blurred' imagery in Google Earth, the satellite images came to us that way from our data providers."

Coverage and Resolution

Some regard Washington as the final authority on satellite-imaging security. "If the U.S. government was deeply concerned, they would have imposed shutter control," GeoEye's Brender says. "I'm surprised -- and pleased," he adds, that they never have.

Google's executives believe that the U.S. has been even more explicit about the subject. The company has issued a statement saying that a Presidential Directive " ... established a policy that favors the public availability of commercial remote imaging data, on the ground that the benefits to the public vastly outweigh the potential risks."

"Presumably," says Pike of GlobalSecurity.org, "the National Geospatial-Intelligence Agency has made some assessment of this, when they did not impose shutter control over Afghanistan and Iraq."

This let-it-be attitude has fueled speculation that the photos online are not military grade, and therefore do not reveal information that only the military possesses or that is otherwise unavailable.

U.S. military satellites are believed to provide real-time or near-real-time streams of images. They are also thought to offer higher resolution, the amount of detail that can actually be seen. From a height of 300 miles, DigitalGlobe snaps photos with a resolution of 0.6 meter per pixel. This means that the smallest detectable image element represents something that is 0.6 meters wide.

Photos from other satellite-imagery companies, such as GeoEye or ImageSat International, range from 0.7 meters to 1 meter per pixel. A next-generation satellite from GeoEye, due to be launched in February 2007, will have a resolution of .41 meters per pixel. And the aerial "bird's eye views" offered on Microsoft Virtual Earth, which are not available everywhere, have a resolution of 6 inches -- or 0.15 meters -- per pixel.

One country, at least, is taking no chances. A 1997 U.S. law requires that imaging of Israel or the West Bank by American-licensed satellites be made available at no better than 2 meters per pixel resolution. While the resolutions used by U.S. military satellite are not made public, it is known that two planned military satellites from Europe -- one French, the other German -- are being designed to provide a 0.5-meter resolution.

The Race To Innovate

Some think that, as the technology gets better and better, governments will have no choice but to step in and impose limits. "I think this is about the end of the road for commercial systems," Pike says. At higher resolutions or greater update frequencies, he says, "[Commercial image providers] would begin to replicate military systems, and the government has no intention of allowing that."

CDI's Hitchens has a different view. The technology behind satellite imaging will get better regardless of any cap governments might impose, she says, so therefore the military is served best by trying to stay ahead of the development curve.

"It's a losing race," she says. There was a time, Hitchens adds, when U.S. commercial vendors had a 1 meter resolution cap, below which they were not allowed to go. It was dropped because foreign vendors pushed the technology toward a higher resolution.

"The more transparency you have among nation states, the more security you have," she says. "And if a nation-state has things they don't want satellites to see, hide them." She points out that the Indian government was able to secretly conduct its nuclear tests without detection because it covered up everything when the satellites were overhead.

The race is not only against technology, but also against the number and kinds of providers. Currently, 13 nations are operating imaging satellites, plus companies in the U.S., Europe, China, Russia, Ukraine, Japan, and India. And the number is increasing.

Onwards and Upwards

With this kind of growth, what will the future hold? Aerial photos can be controlled to some degree, because airspace can be controlled. But outer space is a different matter.

What happens if the barriers to space photography tumble?

In other words, imagine a world in which putting a very high-resolution, real-time imaging satellite into orbit does not involve a hugely expensive, controlled explosion inside a large rocket, a very expensive package, and a complex control room.

Imagine instead that these less-expensive "microsatellites" could be launched by inexpensive vehicles, possibly on a regular schedule, and controlled by someone using a high-powered laptop.

Regular, commercial space flights will likely become a reality at some point. And the history of telecommunications is the history of large, complex equipment eventually becoming smaller, less expensive, easier to use, and more powerful.

"If the ability to launch a satellite becomes so cheap that any company could do it, then all bets are off," Hitchens says. The world would have to set up some kind of space-management program, she says, beyond the patchwork of gentleman's agreements and miscellaneous laws that currently apply to space imaging.

"It creates a whole different paradigm," she says.

Back on the ground, meanwhile, at least one company has managed to evade the all-seeing eye of Google Earth: Google itself.

On the banks of the Columbia River at The Dalles, Oregon, the company is building massive computing facilities in several large, standalone buildings. These facilities, nearing completion, could secure Google's position as an Internet powerhouse for years to come.

But if you search on Google Earth for that particular piece of real estate on the banks of the Columbia River, you won't find these monuments to the power of the Internet.

Instead, you will find a mostly undeveloped piece of land.


Copyright 2006, NewsFactor Network