It is low tide at the Gowanus Canal. The water near 2nd Avenue and 5th Street is dark green and murky. The smell of rotten eggs wafts through the air; it could be from the untreated human waste that regularly empties into the waterway, or from the noisy industry that burdens its banks. This Brooklyn fixture, thick with cancer-causing pollutants, was designated a Superfund site in 2010. But two men head straight for it, cradling a small, yellow robotic boat, like movers carrying a delicate end table. One man wears industrial rubber boots so he can step into the fetid muck without having to fear for his foot.
“Whenever it rains, this is where the raw sewage comes out,” explains Jeffrey Laut, a student at New York University’s Tandon School of Engineering.
Laut and his research partner have reached the tricky part—a rocky path where one loose stone could mean the most awful of submersions. At the edge of the canal, they gingerly place the robot into the water, careful not to graze the surface with their hands. Laut picks up a remote control, pushes a lever, and four lithium polymer batteries inside the three-and-a-half-foot-long contraption start its thrusters, which guide it into the middle of the canal. Several times a month, the machine takes photographs and samples the water quality there. Some days, Laut says, the stench is so bad “you want to vomit.” He occasionally sees dead rats floating in the water. But the robot is impervious to both rats and stink, a mechanical scientist willing to swim where few humans would dare.
At almost 150 years old, the Gowanus Canal represents 1.8 miles of historical neglect, poor planning, and environmental contempt.
At almost 150 years old, the Gowanus Canal represents 1.8 miles of historical neglect, poor planning, and environmental contempt. It was once a bucolic tidal creek prized for its oysters, surrounded by farms and mills. In the 1860s, the then-city of Brooklyn converted it into a canal that soon became a heavily trafficked industrial channel, where various companies began dumping freely into the water. In 1906, two MIT students writing a thesis on the canal called it an “exceedingly obnoxious” “open sewer,” with “foul” waters covered in “oil, coal dust and scum.” Its sludgy banks were in a state of “putrefactive decomposition.”
The canal hasn’t improved much since. A 1998 documentary described how men fishing there discovered a suitcase stuffed with human body parts. After a storm in 2010, spectacular brown waves of sewage billowed over the green water, causing nearby pedestrians to clutch their noses and flee. The Environmental Protection Agency, which has found pesticides, heavy metals, PCBs, and other contaminants in the canal, is currently designing a clean-up job that could last until 2022.
But now, it stands as an opportunity for a small group of NYU scientists. The Brooklyn Atlantis project, as it’s known, was conceived by two professors—Maurizio Porfiri, who studies robotics and applied mathematics, and Oded Nov, who studies interactions between people and machines—after the 2010 Gulf of Mexico oil spill. The idea was to monitor the effects of environmental disaster by harnessing the relative strengths of humans and machines: having robots gather data in environments that are too difficult or dangerous for people, and having people make sense of the data.
“We want the robots to do the hard work, but the people to contribute to the brains,” Porfiri says. The robot has an onboard computer, a 3G broadband connection, and a GPS for mapping the data it gathers. It has two cameras, including one with a panoramic lens. Its photos are uploaded to a website, where about 2,000 citizen-scientists tag them for things like garbage and wildlife. The researchers hope to create a comprehensive picture of the state of the canal, both in real time and over the course of the next few years as clean-up accelerates. They also want to see how the photographic data correlates with water-quality measures such as dissolved oxygen, pH, and temperature, so researchers can track the accuracy of their analysis.
The practical realities of an environmental hazard zone occasionally undermine the researchers’ high ideals. Recently, the team had to put cages around the robot’s thrusters because condoms and plastic bags kept blocking them. On many occasions, Laut, who is also a member of the canal’s canoe club, has had to paddle into the water to rescue his stranded invention. “The water isn’t clean, but you’re not going to die if some gets on you,” Laut says. “A few members of the canoe club have fallen in—they’re all still alive.”
Porfiri, who is originally from Rome, had just moved to Park Slope when his wife pointed out that their friendly neighborhood Superfund site would be the perfect setting for the project he and Nov were dreaming up. With a grant from the National Science Foundation, the NYU team began building a prototype robot and a website, and consulting with the EPA to figure out what to test for in the water. The agency thought the idea was cool, in part because it functioned as community outreach.
Much of the project’s energy is devoted to understanding how to get ordinary folks involved in science—both as a way to boost science literacy and as a way to take advantage of crowd-sourcing efforts that have enhanced fields from ornithology to journalism. It’s way beyond the capacity of a few people to analyze the many images the Brooklyn Atlantis robot takes, which is why the NYU team wants to have many, many people combing through them, repeating each other’s work, to establish what Nov calls “ground truth.” But how do you motivate people to do work for free— especially work as mundane as scouring rows of numbingly similar industrial images for signs of life? The NYU team has run a few studies. In a pilot study of patients afflicted by stroke or cerebral palsy in a children’s hospital in Rome, the team adapted fine-motor rehabilitation exercises by having subjects tag photos of the Gowanus with a haptic joystick, and found that the task increased the patients’ motivation. In the preliminary results of another experiment, subjects who believed other people were doing a lot of work tagging photos were motivated to do more work themselves. Perhaps this information got their competitive juices flowing, or perhaps the comparison “set up norms in an uncertain environment,” Nov says.
It might seem sneaky to use competition to push a virtuous task, but it may also be necessary. To motivate people to do the hard work of change, you have to make that work bite-sized and accessible, and you have to show how their work is helping. On the most basic level, you have to demonstrate that change is possible.
There is reason to be at least a little optimistic about the Gowanus Canal. Laut says the robot’s data has shown higher levels of oxygen in the water near a recently repaired flushing tunnel. And where there is degradation, there is also resilience. Muskrats swim through the waterway. A snowy egret has been spotted surveying the ruined landscape from its perch on a tree above. When the scientists first hauled in their white prototype robot, a heroic crab was sitting aboard, as if to declare all was not lost.
Submit your response to this story to letters@psmag.com. If you would like us to consider your letter for publication, please include your name, city, and state. Letters may be edited for length and clarity, and may be published in any medium.
For more from Pacific Standard, and to support our work, sign up for our free email newsletter and subscribe to our print magazine, where this piece originally appeared. Digital editions are available in the App Store and on Zinio and other platforms.
