MISSOULA — Remote cameras are helping people discover new things about wildlife, but first, biologists must pick through thousands of pictures. A website allows citizen scientists to help without leaving their homes.
Avian biologists Kate Stone and Eric Rasmussen of the MPG Ranch have been on a four-year hunt for the best-looking places in the Bitterroot Valley to dump animal carcasses to attract scavengers.
When they started in the winter of 2016-2017, their targets were eagles, primarily golden eagles. That’s because golden eagles are in decline , and biologists aren’t sure why. But they do know that some eagles settle in the Bitterroot Valley during winter after spending the summer breeding in the Arctic.
To learn how many eagles overwinter, the biologists spread a wide net, setting up about 30 observation sites throughout the valley. Working with the Missoula-based Raptor View Research Institute, they also tagged some of the birds, hoping to see them return over the years.
“This (project) shows the public and the people we work with that something about the Bitterroot is attractive to (eagles),” Stone said. “We want to promote stewardship to make this place the best it can be for them.”
Hauling roadkill bait to 30 sites was hard work, but the real challenge is going through all the photos from each remote camera. Until that’s done, it’s uncertain how many eagles return.
“We had the cameras set to shoot every 30 seconds,” Rasmussen said. “(The first winter) the eagles weren’t tuned in on the sites yet, so 95 percent of the shots were just magpies.”
Deleting magpie photos was easy, but it was tedious and time-consuming for just two people.
That’s when they found Zooniverse , an internet “platform for people-powered research” designed especially for projects that depend on lots of photos. Researchers of all disciplines can create a page, provide instructions for volunteers and upload their photos for volunteers to process.
It has the advantage that volunteers can sign in at any time from anywhere and start clicking on what they see. Some organizations offer incentive gifts – the Bitterroot Valley Winter Eagle Project offered T-shirts – after volunteers process a set number of photos.
“I thought, ‘Cool, that will get people doing this,’” Stone said. “All the sudden, I was sending T-shirts to Portugal, Taiwan and a few places in the U.K. We were shocked at how far away and who was working on our site.”
Zooniverse has the disadvantage that volunteers potentially have a wide variety of backgrounds, and they don’t always read all the instructions. So it’s best to have more than one volunteer assess each photo. Researchers can designate how many volunteers should view a photo before it’s considered complete.
“We set the limit at five people per photo,” Rasmussen said. “It would maybe be better to go with 10 people to get better accuracy. But we didn’t want it to take forever.”
Stone uploaded her first raft of photos to Zooniverse in April 2017, thinking they might be finished by the next season. That didn’t happen. She still has unprocessed photos from that first year but decided to mix it up after discovering how well volunteers work. Or don’t.
The hardest thing is keeping random volunteers engaged. If people have to look at hundreds of photos of the same site containing no animals, they’ll get bored or discouraged. So Rasmussen started shuffling photos of different carcass locations together, especially if the crew set up new sites the following year. He also started trying to pick sites that had more interesting backgrounds.
Once five volunteers have seen a photo, Stone’s biggest challenge is running quality control on volunteers who can’t tell the difference between immature golden and bald eagles. Young bald eagles look brown and haven’t developed their distinctive white head.
“A lot of these species are new to people – even birders have a hard time with eagles. Second to that are the canids – foxes, coyotes and everyone wants things to be wolves,” Stone said. “We have a field guide on the site. But ID tips don’t seem to make an impression.”
Stone said the website has gotten better over the past three years, and researchers like her eventually learn the best way to use it. A few days ago, Stone downloaded the project’s most recent Zooniverse results and was able to log an additional six scavenger sightings, thanks to volunteers.
It’s all helping as the project settles into a busy fourth season. Although eagles are still the main focus, the biologists decided to log data about every animal that visits the carcasses. That means more photos to process, and Stone would love to send T-shirts to a few more Montana volunteers.
“We thought there’s so much to learn about scavengers, scavenger ecology, and who comes to the carcass and in what order,” Stone said. “If we start removing (those photos), then we start missing the whole picture. It’s a balance between getting stuff done and also trying to be accurate about what’s happening in the field.”
The group has streamlined its effort to about a dozen carcass sites that includes about four or five that have been used since the start. They’ve also built up to between 130 and 150 tagged eagles, which is enough to see many returning birds in successive years.
“We have over 100 re-encounters with marked birds as of this year. That’s really crazy and awesome and motivates us to keep doing it,” Stone said. “When you get a bird we haven’t seen since 2016 popping up on some ranch in the Bitterroot and know it’s gone back and forth three times to its breeding grounds and still survives – it’s really great to see.”
Click here to participate in the Bitterroot Valley Eagle Zooniverse project.
Contact reporter Laura Lundquist at email@example.com .