What do you get when you post over 200 cameras across an African savanna? A whole lot of pictures and a tremendous amount of data, all of which must be analyzed and catalogued for use in research. So the scientists conducting the project, Snapshot Serengeti, turned to a reliable work force to sort through the masses by enlisting a dedicated corps of volunteers.

The goal of Snapshot Serengeti was to observe the spatial and temporal dynamics among animals within the Serengeti National Park in Tanzania. They did this by deploying 225 camera traps across the 1,000 square kilometers of park, capturing over 1.2 million images during the first three years of operation (beginning in 2010).

"This was the largest camera tracking survey conducted in science to date," says Alexandra Swanson, a postdoctoral fellow at the University of Oxford who was involved in the project. "We wanted to study how predators and their prey co-existed across a dynamic landscape. We needed to answer different questions than camera traps had answered previously."

But after amassing the photos, they quickly realized they needed a plan for tackling the enormous data set.

"If we were only interested in lions and leopards, we could have classified those images ourselves, but with hundreds of thousands of images of wildebeests and zebras, we simply couldn't keep up with the photos being produced."

Of the million-plus images captured, over 320,000 of them contained animals in the national park. So Swanson and then-fellow graduate student Margaret Kosmala enlisted the aid of a citizen science platform, Zooniverse, to comb through the photos, classifying images, identifying species, and tediously counting the number of animals in each photo. They even characterized animal behavior observed through the many lenses.

"Every image was seen by many volunteers, and we created an algorithm to seek consensus in the identifications," Swanson says. Were it not for the volunteers, the cataloguing would not have been possible.

"Computer vision research is now on the cusp of being able to recognize animals in camera trap images, but when we started Snapshot Serengeti a few years ago, there was no automated way to identify the animals in our pictures. We needed to rely on the human eye."

Fortunately, they had many eyes on which to rely. Over 28,000 registered users contributed 10.8 million classifications, which when processed using the scientists' algorithm, produced final classifications for each image.

"This project is a great example of how citizen science can contribute to real research," Swanson says. "We all know that people are good at pattern recognition, so harnessing the power of volunteers will become increasingly important for ecology studies. We can engage people with no scientific background to help in producing publishable scientific research at a scope and scale that would otherwise have been impossible."

Their research was reported online in the journal Scientific Data. For information about Snapshot Serengeti or to volunteer for similar projects, visit their website.