he end of the earth" is what Scott Madry calls the area of Africa where the countries of Zaire, Rwanda, and Uganda merge and form the Virunga Mountains. "It’s so remote, you have to pay people to carry car batteries up there just for power," he says.

Madry, research associate professor of anthropology, first became acquainted with the Virunga Mountains back in 1993, when he was associate director of the Remote Sensing Center at Rutgers University. Dieter Steklis, a colleague in the anthropology department and a renowned primatologist working under the auspices of the Dian Fossey Gorilla Fund, needed help charting information on the fewer than 600 gorillas left in the area. For instance, he wanted to make a map of the different types of vegetation so that researchers could plot how the gorillas move and what they eat.

Steklis came to Madry not because he knows about gorillas, but because he knows about maps. Sophisticated maps. In fact, Madry, who studied spatial — or landscape — archeology as a graduate student at Carolina in the 1970s, was one of the first handful of scientists to apply geographical information systems (GIS) to archeology.

As a student, Madry says he was interested in archeology but not all that interested in "digging around in a single site for years." He was more into the patterns of ancient settlements or roads, for instance, and showing how cultures interact with the environment over time. For his master’s thesis, he analyzed aerial photographs of France, looking for vestiges of cultural remains — old roadways, foundations of structures — and then plotted them on maps he taped together to see if he could find patterns such as how the layout of roads changed between Celtic and Roman times. As a pilot, he also enjoyed flying around and looking for archeological sites. "Applying GIS to archeology was a perfect fit for me," Madry says.

The idea behind GIS is to take data from many different sources — historic maps, satellite images, aerial photos, field research data — and put it all together so that each set of data aligns with the others — like layers in a cake. "If you were to stick a pin down through it, you’d be in the same place on each layer," Madry says. After all of the data are entered into the system, researchers can look at layers individually or pick two or three to look at together. They can then see the patterns in the data and look for new locations with similar patterns.

or most GIS projects, Madry usually starts with at least some good satellite images and navigable maps. But if you’ve ever seen the film Gorillas in the Mist about Dian Fossey’s efforts to save the mountain gorillas, then you know that the area is constantly shrouded in clouds, making it difficult to get aerial and satellite views of the land.

Compounding the problem is that while the area is one habitat to the gorillas, politically it’s three different countries, each with its own set of natural and man-made strifes. The Rwandan genocides of 1994, for example, caused people to flee the country and set up refugee camps in the Virunga Mountains. Needing food, many refugees set traps. While the people aren’t necessarily after the gorillas, Madry says many gorillas end up getting caught in the traps because a snare can’t tell the difference between a gorilla and a gazelle.

Existing maps weren’t much help. The best Rwandan maps were colonial Belgian maps from the 1930s that Madry found at the national archives in Brussels. Uganda’s most reliable maps were 1950s British maps left over from when Uganda was still a British colony. "But when you put the two together nothing matched up," Madry says. "Not to mention that all of the maps and photographs of Zaire are military secrets and nobody can have access to them."