15 Adorable Animal Selfies That Will Melt Your Heart

START THE DAY HERE

Early voting turnout strong in North Carolina. What Hamas leader’s death may mean for war in Gaza. Investigation into Liam Payne’s fatal fall continues.

Inside Africa

Travel

15 Animal Selfies

Updated 11:49 AM EDT, Fri July 17, 2015

In 2010, researcher Alexandra Swanson set up 225 camera traps across the Serengeti National Park in Tanzania.
In 2010, researcher Alexandra Swanson set up 225 camera traps across the Serengeti National Park in Tanzania.
The traps were set up over 1,000 kilometers of the park.
The traps were set up over 1,000 kilometers of the park.
Swanson wanted to see how different carnivores interacted with lions.
Swanson wanted to see how different carnivores interacted with lions. “I wanted to see how large carnivores share the landscape and how they coexist with lions, because lions interact aggressively with the smaller guys.”
The project is the largest camera tracking survey to date.
The project is the largest camera tracking survey to date.
Swanson collected over 1.2 million animal selfies as part of the project.
Swanson collected over 1.2 million animal selfies as part of the project.
Around 40,000 volunteers helped sort through the image data.
Around 40,000 volunteers helped sort through the image data on Zooniverse, an online platform for citizen science projects.
Cheetahs, hyenas, and jackals share the same landscape with lions.
One thing Swanson discovered was how cheetahs, hyenas, and jackals share the same landscape with lions without getting injured.
Large carnivores treat the area like a 'timeshare'.
She discovered that large carnivores treat the area like a ‘timeshare’, often visiting certain habitat hotspots (those with ample water and shade) in turns.
Combining camera trap data with rainfall data.
“On a much finer scale, we don’t understand how they make their movement decisions. By combining the camera trap data with rainfall data, we are trying to map that migration in more detail,” she explains.
The Zooniverse volunteers classified animals with 97% accuracy.
The Zooniverse volunteers were able to classify animals in the photos with 97% accuracy.
The same image would be sent out to several volunteers for verification.
The same image would be sent out to several volunteers. If everyone correctly identified an animal, it was most likely accurate.
Disagreement about animal identification usually indicates expert review.
If there was disagreement about what type of animal was in a photo, that was usually a good indicator that an expert needed to review the image.


Back To Top