SO THAT'S HOW WE GOT where we are. How one gets to this point in a metaphysical sense is more complicated. "My life's story is pretty convoluted and twisted," Greg told Paolo and me over a dinner of Hawaiian pizza in Tupiza. "But it all kind of relates to exploration."

Greg wanted to go to Mars. This was his earliest dream, and he'd meant it: His majors in college were astronomy and geology. His first foreign language, which he studied during a semester in Moscow, in 1990, was Russian—the era's other language of space exploration. His first real job was as an assistant on NASA's Magellan mission, in the early nineties, examining every photo the Magellan spacecraft sent back from Venus, becoming the first human to "see" large swaths of the planet. His master's was in planetary geology. And his moment of disillusionment came not when he applied to be an astronaut and was rejected—only a handful of the 5,000 applicants made the cut, and he could apply again—but when he realized that modern astronauts were going only as far as the International Space Station.

"My dream was to go on land somewhere," he told us. "I decided I just wanted to explore Earth more." In the grad-school library at Arizona State, he flipped through career books until he found the geophysical firm that worked in the most countries across the globe. That it turned out to be a petroleum-surveying company bothers the environmentalist in him, but Greg's story illustrates how hard a guy has to work these days to find something to explore. He's had to make some sacrifices.

For Greg, the end of the Cold War was a window of real, if fleeting, opportunity. One of his favorite stories is about when the walls were coming down, and he happened to be in Vienna, and he happened to have a raft, and he happened to notice that the Danube River flowed straight into Czechoslovakia. He climbed in and floated to Bratislava. "There was no passport control, and nobody said anything," he recalled. "I just noticed that all the buildings looked different." When he reached the city, he was surrounded by patrol boats with machine guns. When the police realized he was an American—one of the few they had seen—they gave him a hero's welcome, stamping his passport on the spot.

Greg's first geophysical assignment was in the Caspian Sea, which allowed him, during a drunken port call with the mostly Azerbaijani crew, to sneak visa-free into Turkmenistan—a place few Westerners have seen to this day. When the Caspian job was done, in 1999, he and a friend bought a Niva—an old Soviet jeep—and spent months driving it around Georgia and Russia. Siberia was close to China, and China was opening up, so he drifted east, traveling overland until he'd crossed the entire continent. He went to Taiwan, where he became obsessed with learning Mandarin, which he studied until the oil money ran out. "I had to start teaching English," he says.

Globalization kept creeping on, and Greg kept teaching—first in Taiwan, then in Japan. During summers he began leading tours in China for the growing horde of outsiders coming to see it. Asia was becoming less exotic, though Greg himself wasn't. One time he went alone to a Chinese zoo and noticed that everyone was staring at him instead of the monkeys. Trying to lighten up an awkward moment, he hunched over, scratching himself and making ape noises, while the crowd, still expressionless, stared harder.

You might say the confluence project gave Greg newfound purpose. But his brand of modern, confluence-driven exploration poses problems of its own. Our trip to Bolivia, for instance, was originally meant to be a trip to Peru. After I first contacted him, I invited myself along on his next expedition, and we planned it for months—a trek to 12°S 76°W, in the Andes, supposedly the highest confluence in the Western Hemisphere. Then I got a late-night e-mail from him slugged "interesting development." On the DCP Web site, he told me, the Peru confluence had suddenly been demoted to number two, and an obscure point in Bolivia had been elevated to highest in the Americas. The reasons were unclear, and were only slightly less so after Greg's techy explanation:

It looks to me like the project is now
using elevations from Google Earth.
They originally used elevations from the
GT30 1km footprint elevation data.
Then, in 2005, I got a hold of the SRTM
(Space Shuttle Radar Topography) data
(the best data to date), and convinced
them to change the data for the top 50
highest confluence points. Now it has
changed again, and I've already contacted
the project to find out their
source. I want to make sure it is worthy.

Greg went into overdrive to find the source of the updated elevation data, spending weeks e-mailing back and forth with a shadowy Google Earth authority code-named Penguin Opus, a German- and French-speaking Scottish topography expert with an Italian name, and various DCP coordinators in Canada, Russia, and the Middle East. I received messages from him with titles like "russian plot," "a plot of points," and, eventually, "Bolivia." It was finally confirmed: We were going to 18°S 69°W, a confluence that was—according to all the best data sets—at least 300 feet higher than the one in Peru.

A truism of confluence hunting is that you never really know what the obstacles will be. In a string of last-minute e-mails from Brazil, Greg advised Paolo and me to be ready for anything. What looked "so inviting" on Google Earth could be treacherous in real life. We should bring crampons and ice axes. We should bring a tent and a stove. We should factor in extra time for things to go wrong.

"It could be a walk in the park," he wrote, "but, as in a lot of confluence hunting, you just need to be prepared for the unforeseen."