The Wildlife Insights platform will help analyze and share camera trap images, such as this one of a spotted hyena snapped in the Zambezi region of Namibia.
Will Burrard-Lucas/WWF-US
By Eva Frederick
Camera traps—automated cameras that snap a picture whenever an animal walks by—have become an indispensable tool for wildlife biologists, helping them study behavior and estimate populations. But each trap can generate thousands of photos, and researchers often don’t have the time to sort through all the images, pick out their study subjects, and toss the “bycatch”—all the other critters that get their portraits taken. As a result, there are countless “hard drives around the world full of very, very useful data just sitting there, unused,” says Margaret Kinnaird, a wildlife practice leader at the World Wildlife Fund (WWF) in Washington, D.C.
Today, Google Earth, WWF, and other conservation organizations are launching an online database that aims to change that. Wildlife Insights will allow users to upload camera trap images and then have software powered by artificial intelligence analyze them. Users will be able to ask the system to search for their animal of interest, and all of the images will be publicly available. That could be a huge help to researchers, Kinnaird says, saving time and putting a global data set within easy reach.
ScienceInsider talked to Jorge Ahumada, executive director of Wildlife Insights, which is based at the offices of Conservation International in Arlington, Virginia, about how the new platform will work and the impact it might have. The interview has been edited for clarity and brevity.
Q: What problems is Wildlife Insights solving?
A: There is a lot of camera trap data from everywhere, but it’s not being used for conservation. We asked the question, “Why?” And we came up with three barriers. The first is that there’s too much data from each trap for one person to manage. It’s very difficult to get from the raw data to a clean data set because there are thousands of images to look at and keep track of. We wanted to work in a system that would improve that using artificial intelligence, and that’s why we partnered with Google: so the individuals working with camera trap data wouldn’t have to identify all the images by themselves, but a machine would do it first and help simplify the problem and speed up the data flow.
The second barrier that we wanted to improve was having a place where people could put their data and make sure that it’s available, shareable, and people can collaborate. There was no such place. Even if people were successful in organizing all the images from a camera trapping project, they ended up on an external hard drive somewhere. Some might be in cloud accounts, but a lot of these data sets are either in danger of disappearing or siloed. They’re not talking to each other. The only way to bring that together was in this single platform in the cloud.
The third barrier is that even if you speed up the processing of data and [you] can share it, there’s still a lot of barriers in understanding what the data is telling you. You need a lot of technical knowledge and training to be able to make a camera trap data set and calculate how species in that data set are changing through time or what the species richness in the sample is or basic statistics like that. That requires a lot of work that most people don’t have the skills to do. So, our solution was to create an analysis module that will automate all that analysis.
Q: How will this platform change the way we study animals in the wild?
A: Wildlife Insights will basically give us information that we don’t have access to at all. For many animals, we don’t really know how their population is changing or how many animals are in the population. Those are basic facts that you need in order to manage a conserved species that we don’t have for most animals on the planet. Wildlife Insights will give us that for hundreds of species.
The other thing I would say is that this data will help reveal a lot of the inner lives of animals. For example, how do animals use their time during the day? Are they moving, eating, sleeping? That helps us understand changes in the activity patterns of animals that usually happen when there’s some change in the environment, either because of climate change or because of people hunting them. There was a paper in Science a few months ago, using camera trap data, about how animals have become more nocturnal because of people. Camera traps can help us document and understand things like that. And as climate change continues, a lot of animals will have to change their activity. [They could] move into colder periods of the day or nighttime just because of heat stress or things like that.
Q: Will the platform help people other than scientists?
A: Yes, actually, a lot of our potential users, I hope, will be local and Indigenous communities. A lot of these people live in remote areas. They have access to biodiversity and they sometimes use wildlife as part of their livelihoods. They don’t normally know the impact they’re having, and they would like to know. That’s one user group that we’re very interested in getting to.
Another group that I’m very interested in and very passionate about is citizen scientists. That is a very diverse group of people, from nature lovers who just want to put a camera trap in the property to see what they have, to wildlife-savvy land managers and hunters that are interested in knowing how the species there are faring and how to manage different species like deer or turkey. Camera trapping is not a citizen science activity now, and I want it to become one. I think the potential is enormous and it’s a lot of fun for education.
I live in Takoma Park in Maryland, and I put a camera trap in my backyard every winter. I am now analyzing the data, and it’s amazing what you get in your backyard. I get deer and foxes and possums and squirrels—there’s a lot of wildlife. Imagine if every kid in Takoma Park had a camera trap and put it in their backyard. We could put the data in Wildlife Insights and then kids would be doing little competitions to find out who’s got more species in their backyard—and they would learn about wildlife. It could be a fantastic tool that really has the potential to bring people together and rally them around conservation.
Source: Science Mag