Geographic Information Systems Take Off

** I wrote this in December & never posted it **

When the Russians launched Sputnik into space 50 years ago, the age of spy satellites began. In response, the US deployed surveillance satellites to observe Soviet military operations, monitoring their production of long-range bombers and ballistic missiles. These satellites dropped black and white film canisters to earth in parachute-equipped capsules, so planes could recover them in mid-air. Analysts with high level security clearances would pore over the film, identifying potential threats.

No longer cloaked in secrecy, satellite imagery falls in the public domain these days. Transmitted digitally and parachute-free, satellite images - once reserved for the eyes of CIA agents - are now available to anyone. Computer users can download software providing images of destroyed villages in Burma and Darfur, or a suspected Syrian nuclear site which was bombed by Israel in September – with a before and after shot available. Similarly, one is able to see the town of Greenburg, Kansas – as it existed before and after a tornado flattened it in May 2007. Those exploring the Dutch city of the Hague may even come across topless sunbathers, who should’ve known that they were exposed – even from outer space.

A bird’s eye view of the world is now available to anyone for free. Using services like Google Earth, Microsoft Visual Earth and Yahoo Maps, which combine high resolution satellite and aerial imagery with electronic maps, web surfers can see almost anything from above. While these geographic information systems have become a source of fascination for the public, security-conscious government officials remain wary and distrustful. They worry that these systems have made their countries more vulnerable than ever.

Satellite images went from guarded state secret to sellable commodity during the early 1990s, when private sector companies first gained permission to collect high-resolution images of Earth and sell them to any consumer – be it a foreign government, private citizen or company. Congress passed the Land Remote Sensing Policy Act in 1992, allowing the Secretary of Commerce to issue licenses to private companies for space-based remote sensing systems. Private companies could finally launch satellites collecting data about the surface of the Earth.

The regulatory framework had taken shape, but the commercial satellite imagery industry would arrive years later. In September 1999, Space Imaging launched the first private satellite. Other satellite launches by the private sector would soon follow. To date, more than two dozen companies have received licenses to operate in space, and more than 45 satellite remote sensing systems will be launched over the next three years.

Surprisingly, the federal government serves as a major customer for this emerging private sector. Even though the federal government has its own satellites to draw upon, its agencies actively acquire commercial satellite imagery as well – almost perceived as a subsidy and show of support for a nascent industry. The National Geospatial-Intelligence Agency will spend up to $500 million over five years, acquiring commercial imagery on behalf of a number of federal agencies.

While intelligence agencies and other government agencies have purchased images from the private sector, an emerging business sector has also taken interest in acquiring imagery. A company called Keyhole was founded in 2001 and set out to deliver a 3D digital model of earth by acquiring satellite images and stitching them together. Google acquired Keyhole in 2004 and relaunched its product as Google Earth in 2005. With a powerful brand behind the technology, usage of geographic information systems proliferated.

Geographic information systems carry a certain ‘wow’ factor – perhaps explaining why Google Earth has been downloaded over 200 million times since its debut. A Google Earth user can show a friend where he was raised, zooming in on his childhood home, elementary school, and the baseball diamond where he played Little League. A user can even activate layers showing where restaurants, banks, gas stations, hospitals and shopping areas are located – in case anyone was wondering how many Starbucks can be found within a ten block radius.

With all of this powerful information at users’ fingertips, many people in the security business feel threatened by a world with less secrecy, a world where anything is visible. Operators of Australia’s Lucas Heights nuclear reactor asked Google to censor photographs of the site in late 2005, expressing concern that the images could be used by terrorists. In January 2007, the Daily Telegraph reported that terrorists were using Google Earth to aid in attacks on British bases in the Basra region of Iraq. Raids on insurgents’ homes revealed print-outs of images from Google Earth – photographs showing vulnerable spots within bases, such as tented accommodations, lavatory blocks and lots full of lightly-armored Land Rovers. After military officials complained, Google Earth posted older pre-war images of Basra, replacing those which showed sensitive military installations.

Nothing in international law compels Google Earth to act on such requests to censor material. Joanne Gabrynowicz, a law professor at the University of Mississippi and director of the National Center for Remote Sensing, Air and Space Law, says that the company will typically field complaints by a foreign government or entity regarding a sensitive site and look at them case-by case. The decision over whether to blur images or replace them with older ones hinges upon “the country it’s coming from, the foreign policy relationships involved, and what arguments they make,” says Gabrynowicz. “You’re not going to find any clear, uniform rules.”

Most of the changes made by services like Google Earth and Microsoft Visual Earth are done quietly – although bloggers remain vigilant and ready to criticize any intentional blurring or reversion to older imagery of a given site. For example, Google faced intense criticism when it replaced its post-Katrina imagery of New Orleans with shots of the city before the hurricane struck.

While some suspicious changes have been made, censorship and distortion of satellite imagery has not been as brazen as one would expect. “I think it doesn’t happen often, and not intentionally,” says Stefan Geens, who runs the popular blog Ogle Earth and tracks Google Earth developments. Earlier this year, The Times of India reported that Google Earth had agreed to use “fuzzy, low resolution pictures and distorted building plans” to protect sensitive military and scientific sites in India. Google Earth has denied that such sites were deliberately distorted, while noting that imagery from its data suppliers sometimes happens to be blurred or degraded – for example, pictures of Vice President Cheney’s residence in Washington, DC, which were blurred but by not by Google’s doing. Rather, the company Google purchased the imagery from could be held accountable. Geens says that contrary to popular belief, most Google updates have only yielded improvements. “I’ve never seen the resolution lowered after an update – it’s always gone to higher resolution,” says Geens. “The Basra switch was likely done at the behest of Allied forces, but the New Orleans update was just a well-meaning update to better quality imagery, albeit older (pre-Katrina). And it was quickly replaced when criticism poured in.”

The US Department of Defense has made little attempt to suppress satellite imagery – even when alarm bells sound. Earlier this year, a Microsoft Visual Earth user stumbled upon a photograph of a propeller on an Ohio-class ballistic missile submarine, which was docked at a maintenance facility in Bangor, Washington. The propeller had been considered a top secret design feature, allowing the submarine to remain quiet and stealthy in the water, avoiding detection. “The Navy had been really good about keeping their propellers under wraps – for decades,” says John Pike, director of the defense think tank GlobalSecurity.org. “They had been really good about that and then they got careless.”

Pike notes that services like Visual Earth and Google Earth typically raise little concern for domestic authorities because they show satellite and aerial images which “have already passed muster in terms of public releaseability.” The images on these services have been purchased from commercial satellite imagery firms, like DigitalGlobe and GeoEye, which received the go-ahead from the federal government to sell them.

Geographic information systems present a big hodge-podge of data, collected at all different times and acquired from many different sources. Images do not bear a time stamp and are not in real-time. In fact, images can range from two to four years old. Only in select instances – like the Greensburg, Kansas tornado case – does more recent imagery become available.

“Because a lot of the imagery is relatively old, you would have to question how much national security is being compromised,” says Gabrynowicz. “The companies themselves – depending on who asks them to withdraw or modify imagery – are making business decisions. They’re not making civil rights decisions or 4th Amendment decisions.”

Despite the age of the imagery being displayed, many professionals find geographic information systems useful and see them advancing their fields. UNC-Chapel Hill archaeologist Scott Madry hopped on Google Earth two years ago and used the software to find over 20 unique archaeological sites in France. Madry looked for signs of structures, roadways, and fortifications, and within 10 minutes, he was making discoveries. “I found more stuff in my first week on Google Earth than I had in 20 years,” says Madry, who has spent hundreds of hours in aircraft, canvassing Western Europe for ruins. Like his fellow global citizens, Madry prefers this new approach where he doesn’t have to leave his office in order to see the wider world. “It’s just an astounding tool. It has tremendous potential for increasing the productivity of how we do this type of aerial archaeology,” says Madry.

The applications for geographic information systems are almost limitless – and there are still opportunities to optimize these systems. In 1998, former US vice president Al Gore envisioned a “digital earth,” one in which a virtual, 3-D representation of the planet that is spatially referenced would be connected to the world’s digital archives. Following Gore’s vision, programs like Google Earth would not just show you what a place looks like – they would also tell you about its history, its people, its climate. “You could point to a location on Earth, but then find other related information – about hydrology, geology. What is the kind of rock my house is built upon, what type of soil,” says Professor Keith Clarke, a GIS expert at the University of California Santa Barbara. Clarke adds, “And there’s an enormous amount of demographic and social data available that’s not yet linked through geobrowsers (like Google Earth).”