How crowdsourced data helps cities fight litter
Analyzing the amounts and composition of litter, combined with information on location, transit patterns and days of the week, is giving cities insights for better intervention and messaging strategies.
To combat its litter problem, the city of Memphis is turning to data.
Over four evaluation sessions this year, 10 researchers will monitor whether litter-prevention methods are working as part of the City Fingerprint Project, a three-city initiative by data science company Litterati to study litter-related pain points and use data to drive solutions. Hayward, Calif., and Norfolk, Va., are also participating.
For the project, citizen science researchers who are trained by Litterati take a photo of litter items with the company's Researcher app, which geocodes. Later in the process, the images are assigned a tag to identify them. As data is collected by researchers, cities can see what kind of litter is showing up where.
Cities participating in the Fingerprint Project have the additional ability to overlay different datasets – of population density or commercial corridors, for example -- to learn “where the litter could be coming from so that we can make better intervention and messaging strategies,” said Janet Boscarino, executive director of Clean Memphis, a nonprofit environmental protection organization.
The work will expand on Litterati’s Analyze platform, which applies machine learning to photos of litter to develop a baseline of litter composition, such as material, object and brand. The project will look at litter pileup in various sections of the city and compare what it looks like in, say, school zones vs. transportation hubs to give cities a better sense of what litter is most problematic and where.
Memphis has provided several datasets to Litterati on traffic, transit stops and other geospatial data to overlay on their data collection.
“It’s really about getting a better understanding of where our pain points are, what the materials are in those pain points, and what kind of connections there could be to these other sets of data that we’re about to be able to evaluate with this platform to help drive better decision-making,” Boscarino said.
Historically, litter tracking was a largely manual process based on the quantity, not composition of litter. But saying that a city collected 3,000 bags of litter vs. 2,000 bags last year isn’t necessarily meaningful, Litterati founder and CEO Jeff Kirschner said.
“Our belief is we don’t see change because there is no accurate platform for measuring and monitoring the change,” Kirschner said. That’s because the granularity of data has not been available until now – meaning the ability to integrate litter amounts and composition with mobility, days of the week and times of the day to see how they affect what’s on the ground.
Data on litter composition is crucial, said Lisa Brown, project manager at Memphis Transformed, a campaign aimed at creating a zero net waste circular economy. “That informs us about potential source-reduction campaigns that we could potentially run within our city or neighborhood,” she said. “There are a lot of different strategies and intervention methods that we can test out by having this data now.”
The data can also help the Public Works Division and Clean Memphis better deploy worker and capital resources because they can target litter hot spots or put out more trash receptacles.
Litterati issued a request for participation in the Fingerprint Project last summer, and in November 2021, work kicked off in the cities, which the company selected based on geographic dispersion and willingness to share data and access to city workers. They’re preparing to do the first data collection using a consistent protocol and methodology over a stratified sampling of 600 100-meter segments, or nearly 40 miles.
“The primary benefit is to learn how to solve a city’s litter problem. We call it customer discovery,” Kirschner said. “It’s not important what you think the solution is. [It’s] do your customers, do your clients, do your constituents think this is solving their problem?”
When the project ends, the participating cities will get a report that they can use to make positive environmental changes, and Litterati will be able to expand Analyze with nationwide data.
“You could have a litter indexing, where cities who have common characteristics – maybe it’s that they’re all the same population, maybe it’s that they’re all in more temperate zones, maybe it’s that they’re all in very cold areas – could share with one another,” Kirschner said. For instance, cities where recycling bins are open could learn from those where they’re closed.
What’s more, cities could use the national data for inspiration. “It’s always useful to understand what your city is facing and how that compares to other cities,” Boscarino said. “I also think it helps us as a city feel like we’re part of something bigger – a bigger movement, a bigger strategy.”
And that can motivate more people to get involved. For instance, Memphis has used Litterati’s Engage app since 2020 to encourage residents to pick up trash through several challenges, such as the Earth Day 30 Days Straight.
Memphis earned the Ernest T. Trigg “Nation’s Cleanest City” award in 1948, 1949, 1950 and 1951, and its City Beautiful Commission received eight national awards from Keep American Beautiful between 1980 and 2002. But like many cities, litter is a growing problem. In fact, the “Keep America Beautiful 2020 National Litter Study” found that litter increased during the COVID-19 pandemic, with 207 million pieces of personal protective equipment like masks or gloves on the ground at any given time in 2020.
This story was changed March 22 to clarify the litter data collection process.
Stephanie Kanowitz is a freelance writer based in northern Virginia.
NEXT STORY: City makes cross-agency data sharing easier