Three years ago, Shaun Winterton was looking at photos of bugs on Flickr. Winterton is an expert on a type of insect called lacewings—beautiful, strange creatures with long, translucent (and, yes, lace-like) wings. There are more than 13,000 photos on Flickr with "lacewing" in the description, but one in particular caught Winterton’s attention. It was a photo from Malaysia, taken by a man named Guek Hock Ping, and it captured an animal with unusual blue spots on its wings.
Winterton contacted Guek, and asked whether he could go and take more photos of the bug—but it wasn’t until a year later that Guek was able to find another specimen. When he did, however, Winterton realized that he was looking at a previously undiscovered species of insect. A year later, Winterton and Guek published a paper together, describing Semachrysa jade.
Their finding points to a new kind of naturalism—one aided by the computers so many of us carry with us as we live our lives. With millions of cameras and smartphones all over the planet pointed at the natural world, the chances that somebody might catch a new species or an unknown behavior have skyrocketed. And scientists are increasingly trying to tap into that vast pool of information. Wading into the billions of photos and videos shared on social media isn’t easy—but if researchers can harness our cameras, they may be able to unlock a huge amount of information about our world.
Take that video of ants making a daisy chain to pull a millipede that, last week, had scientists scratching their heads. The footage showed a behavior that entomologists hadn’t seen before. And it was shot by an amateur somewhere in Southeast Asia.
But, then, it's not just new species or behaviors that smartphones are documenting. Sometimes a photograph with a geotagged location might be the only record of a species in that location. "I’ve come across blog posts that are making new geographic records, I’ve had tweets that are the first and currently only published records of some species," says Morgan Jackson, an entomologist at the University of Guelph.
These are the kinds of things that citizen science advocates dream of: collaborations between amateurs and trained scientists that produce real, published results. But citizen science projects are really hard to design—they have to be fun, interesting, and not too difficult for participants, while also being robust and scientifically useful. Winterton’s discovery of a new lacewing species is almost a serendipitous citizen science project—one that emerged not by design, but because a human had trawled Flickr and noticed something odd.
But how do you replicate that? There are more than eight million photos on Flickr tagged with the word "insect." Even if you try to drill down to things like "butterfly," "spider," or "beetle," you’re left with a crushing wall of photos, most of which aren’t going to be useful. That’s just Flickr. There is almost certainly a ton of useful data squirreled away in tweets and Facebook albums and YouTube videos all over the place. It’s finding it that’s the problem.
"You know how hard it is to find a certain Tweet—so trying to find one you don’t know exists is nearly impossible," Jackson says.
Here’s an example. This weekend, a strange-looking spider perched itself on the bow of my kayak. When we stopped on the side of the river for a snack, I snapped a photo of my spindly masthead and sent it off to Twitter, asking: "Bug people, what is this thing?" Within a few minutes I was informed that it was a Tetragnathid, also known as a "long-jawed spider." If you search Instagram, there is even a tag for "tetragnathidae," with 28 photos in it.
"Some of those aren’t tetragnathids, but that’s okay," Jackson says. He’s just happy people are tagging bugs more specifically than I did by saying "what is this thing."
Jackson thinks that if scientists can teach people just a little bit more about bugs—or any animal or plant, really—users could help filter some of the pictures they are taking. In his ideal world, I would know enough about spiders to be able to identify the "thing" on my kayak to the family level—tetragnathidae, in this case—and tag it for scientists to find.
It’s not just users that have some learning to do to make their photos helpful, though. Scientists also have to think to look to social media for information. "Not many taxonomists that I know would think to look through Facebook or Twitter hashtags," Jackson says, "even though they wouldn’t think twice to fly halfway across the world to sit in a foreign museum for a week to do essentially the exact same thing."
Using data scraped from users also raises interesting ethical questions. "You took that photo with your cell phone, so it has GPS embedded into the meta information," Jackson says. "I know the time, date, and who took it and where. There’s no reason I can’t use that in an information map. But I don’t know whether that’s unethical to do without that person’s permission."
Nicholas Evans, who works on research ethics and information technology at the University of Pennsylvania, says that these are questions the field will have to grapple with if scientists want to start using this kind of data in published papers. "Say we’re trawling through Twitter looking to make a map," Evans says—"then we can expect that data will be reasonably anonymized. We’ll just use the location information and nothing else. But if we’re using really small amounts of data, we’re talking about revealing where people live, depending on where they found it, and that’s different. We’ve learned that it’s a really bad idea to broadcast where you live on the Internet."
If you took that photo in your backyard, in other words, anybody who reads that paper knows where your backyard is.
There’s also the issue of crediting amateurs who may have been instrumental in a scientific discovery. In the lacewing case, Winterton and Guek were coauthors on the paper. In other cases, a scientist might name the new species after whoever snapped the photo. I asked Jackson if we’re about to enter an era of species with names like "frogsplaysoccer145" or "h0ckeymom."
"If all I’ve got is a username, and I can’t get in touch with them, personally I would consider it," he says. "It would get me some side-eye, but I wouldn’t have a problem if that’s the proper credit to put there."
And storing all this data will be crucial, too, another place where privacy and science might butt heads. If someone deletes his or her Instagram or Twitter account, and makes the conscious choice to remove that information from the web, can researchers still hang on to an archived copy?
The solution might be as simple as contacting the photographer and working out what she's comfortable with, like Winterton did with Guek. But in cases where you’re trying to use thousands of images, that can get unwieldy. And in the case of the ants, where the original footage has been reposted by someone else, finding the person who first captured an image or shot a video can be difficult.
As with almost everything, the field will likely struggle to keep up with technology. "Taxonomists work on the order of decades to almost centuries in some cases," Jackson says. "So to have things like this where you could have a new data set every day to be testing your ideas and hypotheses with—it’s outside the scope of what we’re used to working with."
If they can figure it out, though, researchers could be on the verge of tapping into an incredibly valuable resource. There are more than 500 million photos uploaded and shared every day. Some of those photos are bound to depict things nobody has ever seen before. It's just a matter of finding them.
Read The Coming Age of the Internet Naturalist on theatlantic.com
More From The Atlantic