May 9th, 2023
Courtesy of The Washington Post, a look at how naturalist apps, designed and managed by scientists with world-class data, help us reconnect with the world around us:
First I see the wall barley, like tiny fields of wheat on the side of the road. Then a profusion of musk stork’s-bill overflowing with purple flowers. That’s just the crack in the sidewalk.
By the time I reach my office, I’ve identified dozens of species, most unknown to me a few hours earlier.
I’m not a master naturalist, but I have one in my pocket. Thanks to artificial intelligence trained on millions of observations, anyone with a smartphone can snap a picture or record a sound to identify tens of thousands of species, from field bluebells to native bumblebees.
If I’m honest, it’s the kind of thing I would normally miss while walking or pedaling to work. Birdsong might be gorgeous but I’d barely hear it. I’d note “pine tree” as a catchall for conifers.
That has changed. I’m now on a first-name basis with most of my wild neighbors. It has reconnected me to a natural world I love, yet never studied deeply enough to know all its characters and settings. I’m hardly an exception.
Homo sapiens sapiens may have never been this divorced from nature. Humans across much of the industrialized world have become an indoor species: About 90 percent of our time is spent inside.
It’s hard to fathom. For 2.5 million years, humans spent a huge share, if not virtually all, of their time outdoors. Today, many adults are spending more hours on screens than outside.
Nature is even disappearing from our books, songs and culture, say researchers who looked at nature-related words in popular works during the mid-20th century. Our mental and physical health has declined alongside our estrangement from the outdoors.
But what technology has torn asunder, perhaps it can begin to mend. For people who don’t know the difference between a robin and a magpie, this new generation of naturalist apps is the Rosetta Stone to the natural world. Reestablishing relationships with your outdoor neighbors might not only transform your commute, it might change your life.
Four apps to rule them all
There are more than a dozen apps promising to help you identify the natural world, many of them paid. Don’t bother. Four apps, designed and managed by scientists with world-class data, meet all your ID needs free of charge. And every observation will advance our scientific understanding of the natural world.
The easiest to use is Seek. The app, an offshoot of iNaturalist, a joint initiative of the California Academy of Sciences and the National Geographic Society, lets you shoot live video. It automatically grabs frames and analyzes them. The augmented reality experience is like downloading a foreign language into your brain. The app identifies the taxonomy of plants and animals instantly as you shoot. If it can’t figure out the species, it will give you its best guess.
In less than an hour, I had racked up dozens of plants and insects near my house from Bombus vosnesenskii, a native yellow-faced bumblebee, to the purple-flowered bush lupine it was buzzing around. The only drawback? The app doesn’t include deeper context about the species it identifies.
For that, there’s iNaturalist and Pl@ntNet. Both offer sophisticated, if slightly less user-friendly, apps that upload and analyze photographs of flora. In seconds, they typically return a ranked list of potential candidates with rich descriptions of each. The identification of the most common species is a slam dunk. For rarer ones, it’s easy to compare your observation against those of others in the database.
The apps’ real superpower is the community around them: Millions of citizen scientists who can vet and confirm your observations. It’s particularly satisfying to watch your skills — and ranking — rise in the apps as you get to know your neighborhood. When you’re ready to up your game, download these apps.
Finally, there’s Merlin Bird ID, a project of the Cornell Lab of Ornithology. Merlin feels like magic. The app uses a phone’s sensitive microphone to identify bird vocalizations in the sonic landscape around you, painting a visual representation or sonogram analogous to a musical score.
Within seconds as I’m walking through San Francisco’s Golden Gate Park, the app recognizes the high-pitched staccato of a dark-eyed junco, as well as the falsetto of the Pacific wren.
Merlin has permanently changed how I hear the world. I can now tune in to birdsong operas that had never entered my consciousness. Within a day, I was able to recognize distinct calls without consulting the app.
That, of course, was the point, says Grant Van Horn, a machine learning researcher at the Cornell Lab, who helped build Merlin’s sound ID feature. “If you ever go on a bird walk with someone who knows their birds, it is crazy cool,” he says. “Our inspiration is getting that expertise and sharing it with anyone who has a phone in their pocket.” They succeeded.
Advancing science
But the apps are more than tools to get acquainted with nature. They’re pushing AI identification — and conservation — forward. Recognizing natural inhabitants, and our relationship to them, helps us rediscover what remains and protect it.“We’re just at the beginning of actual real scientific progress,” says Van Horn. “And none of this stuff happens without a passionate group of people that helps you curate, train and evaluate the data. There’s still a ton of opportunity for these passionate communities to contribute their expertise.”
The first big breakthrough came around 2018 from Snapshot Serengeti, a research project using digital camera traps to photograph thousands of migrating African animals. Organizing this enormous collection featuring a variety of animals, from wildebeests to giraffes, proved overwhelming for the small team of scientists.
So researchers enlisted thousands of online volunteers to sort and label more than 3 million images. That allowed Jeff Clune, then a computer scientist at the University of Wyoming, and his collaborators to unleash algorithms on what was at the time the world’s largest collection of labeled wildlife images. The new algorithms could identify animals in 99 percent of images with the same accuracy as human volunteers, around 97 percent, according to a seminal paper published in Proceedings of the National Academy of Sciences. When applied to all data, it could save an estimated 8.4 years of human labor.
Breakthroughs like this are why the apps in your hand can now identify daisies, dandelions and, if you are on the African plains, a lion, Panthera leo. And every observation you contribute makes these a little bit better.
Citizen science-powered algorithms are now going beyond individual organisms. They’re mapping their relationships to an entire ecosystem, from the flower a butterfly pollinates to the leaf where the insect lays its eggs.
“My goal is to turn ecosystems into fire hoses of data,” says Clune. “In the same way a video game company knows everything that happens inside their system, we should know that for the Amazon rainforest. Imagine what that would mean for science. We could answer questions we would never have been able to do.”
Naming nature
Ultimately, the apps’ greatest breakthrough may not be technological at all. It may be raising our awareness. We are nearly blind to entire categories of living creatures. In her book Braiding Sweetgrass, Robin Wall Kimmerer described it as “being lost in a foreign city where you can’t read the street signs,” a form of species loneliness. While these plants and animals are our neighbors, we scarcely acknowledge their existence, let alone their right to exist.
By naming my wild neighbors, I’ve found my perception of them transformed from grainy and distant to powerful and familiar.
Author Jenny Odell writes about a similar experience in the book How to Do Nothing: Resisting the Attention Economy. The simple act of paying attention to the birds around her home in Oakland, Calif., led her down a path of reclaiming her attention from the frenetic, exploitative digital cacophony.
Now, instead of asking what’s there, she asks who: a raven, robin, song sparrow or nuthatch. Instead of a blur of green, she sees redwoods, oaks and blackberries.
This could reverse one of the great losses of the past century: our severed connection to the unique, wild character of where we live.
,