Landsat on a gold mine
To put a spin on the last MapLab lede: If we didn’t know what the world looked like, the world would look pretty different.
That’s the thrust of Abhishek Nagaraj’s research. A professor at UC Berkeley’s Haas School of Business, Nagaraj devotes himself to the question of how maps not only reflect the world but also shape it, in terms of human behavior, political outcomes, and his speciality, economics. To earn his Ph.D., Nagaraj showed how the arrival of Landsat transformed the geographic and competitive landscapes of the gold mining industry, which has long depended on maps to target terrain for extraction.
Launched by NASA in 1972, Landsat is the satellite imaging program designed to map the entire surface of the earth, which it eventually did. But in the early years, different regions of the planet were captured by its orbiting eyes at a somewhat staggered pace.
In his dissertation, Nagaraj took the regions of the planet that were most extensively mapped in Landsat’s first 15 years and compared them to a private, hand-collected database of major gold discoveries dating back to the 1950s. Controlling for variables like the price of gold and the level of financial support for mining in each country, he found that Landsat maps doubled the likelihood of new gold discoveries in the areas they covered.
Part of the explanation, Nagaraj found, was that Landsat provided unprecedented public access to geographic knowledge. That helped mining startups compete with more established firms. Before, aerial images of the planet had to be taken by aircraft, an expensive proposition that only larger companies could afford in limited doses.
Not everyone will be thrilled to discover how extractive industries take advantage of government maps to do their work, environmentally devastating as it frequently is. But that’s the thing about information in the public domain—the U.S., at least, hasn’t generally put many limits on what can be done with it. But, on the Earth protection front, Landsat imagery is equally instrumental for climate researchers and first responders to fires, floods, and other disasters. As MapLab pondered earlier this month, how much more of the West would be burning if not for these views from space?
“Maps have a local impact,” Nagaraj told me earlier this summer. “It’s a story that plays out over and over again on a global scale. And don’t even get me started on interplanetary maps.”
Oh, I will get him started. In a future issue.
Axios reports that the Russell Senate Office Building near the U.S. Capitol has already been renamed the “McCain Senate Office Building” on Google Maps, a few days after Senator Chuck Schumer proposed the change in the wake of Senator John McCain’s death. Given that the Senate hasn’t agreed to it, the new designation is a bit premature.
But questionable labels on Google Maps do actually shift how people talk about neighborhoods—another example of how maps create the world in their image. Last week, I asked you to send in local examples of where Google Maps gets it wrong, or at least doesn’t match up to common parlance. I got an earful. (Some of you asked how to report issues on the map: here’s the link.)
Two examples stood out, because they suggest that Google Maps isn’t just borrowing from real estate developers to rename neighborhoods. It seems it’s also digging deep into history.
From Austin, Texas, Zack Lofton sent in the mystery of a neighborhood called “Abercrombie,” which he said he and his colleagues were laughing about recently:
We’ve never heard the neighborhood referenced with this name, which is funny because we’re all planners with a special interest in this sort of thing. Apparently, this name came from a train station/old town settlement many years ago.
Writing from Berlin, Guglielmo Reina pointed out how a bit of pre-war history crops up on one of Google Maps’s labels in his native city, which “95% of the people living in Milan wouldn’t know what you’re talking about if you mentioned.” But he’s generous about it:
See that Bottonuto in the middle of the city? I grew up in downtown Milan, lived there for 21 years, and am not completely oblivious to Milan’s history; still, I found out about this name looking at Google Maps many years ago. It turns out Bottonuto was a neighborhood in Milan, until it was demolished during the Fascist era.
Naming neighborhoods is hard, and rather than have no name for that area Google decided to give it its old, fallen-into-oblivion name. It’s not completely random, from their point of view, and it’s not factually wrong, so be it.
Google probably sees a business case for making the de facto map of the world as information-packed as possible, which would include throwing in historic place names. Having maximum data, after all, is Google’s M.O. I’ll share more on this when I hear back from their spokespeople. If you’ve got thoughts, drop me a note.
Startographer in the making: a four-year-old has caught a longstanding error on the Washington, D.C., metro map. Trolley problems: the American streetcar lines that disappeared in the late 1950s. F that noise: the New York City Department of Buildings is now mapping active construction projects. Urban violence in the U.S. is on a long-term decline: see where. Digital redlining: That’s what critics say some of Arkansas’s new Medicaid requirements amount to. Pin-drop zero: A researcher is studying how maps on the web go viral. Speaking of satellites: Here goes the world’s first wind-mapping orbiter.
One month left of summer, many more months of MapLab. Tell your friends to sign up for this newsletter. See you in September!