Why do some cameras look at the road?

What's with those two or three cameras that seem to ONLY take pictures of vehicles - or even worse, of nothing at all? If you've clicked through more than a couple of images on Eyes on the Wild, you know the cameras I'm talking about. The one that looks at the birch stand, and the one overlooking an open field with cars in the background. Sometimes, the one where the game trail runs out through a little wooden gate to the road. Maybe you're one of the many folks who have posted good questions about the purpose of these cameras to the project, or complained about how often they pop up. Or maybe (like me) you just click through them, feel thankful that there aren't any extra questions to answer, and move on to the next photo hoping for something more exciting. Either way, we're here with an explanation! 

Which of these three camera spots has caused YOU the most headaches and frustration?

In most trail camera projects, cameras are situated to look at areas where certain types of animals congregate or pass through regularly. This might be a watering hole or game trail on the African savanna in a project like Snapshot Serengeti, or a heavily-used nesting area like in this Least Tern project. Other projects design their camera placements to answer particular questions: Do watering holes concentrate parasites and increase disease transmission? Where is the ideal place to build a wildlife overpass and prevent animals from being hit by cars? Depending on what you are hoping to learn, there are many different ways you can deploy your cameras. 

Because we are interested not just in a single species, but in the entire ecosystem and the interactions that are occurring in it, and because our habitats at Cedar Creek are extremely patchy, we opted for a fixed grid design. Instead of trying to focus on game trails or other areas that we knew certain species were using, we placed cameras at fixed GPS locations and compass bearings on a grid spanning the whole 5500 acre property. This essentially gives us eyes on all our wildlife and all facets of our landscape, whether they are marshy areas (inaccessible to terrestrial mammals except when frozen in the winter, but used by aquatic and semi-aquatic species like mink, beaver and waterfowl the rest of the year) or forests (loved by deer and bear and coyotes, but scorned by open-landscape species like pheasants), highly secluded spots (used by wolves and foxes) or areas near people and development (look for squirrels, raccoons and the ever-present deer). As it turns out, a few of these grid locations look at roads.

Since the reserve is a patch of open space in an increasingly developed area, cars are a fact of life for our wildlife and change how they see and use the landscape. If you look at the below map of Cedar Creek, you'll see a major county road cutting the property in half - that's what the birch stand camera near the dot marked Headquarters is looking at! 

As you might imagine, the presence of a major road can dramatically impact wildlife and their behavior. They may choose to never cross the road, and thus cut themselves off from good habitat on the other side. They may change the time of day that they are active in a particular area, crossing the road only at night when there are fewer cars. Even though it doesn't seem like it because of the high number of cars captured by those particular cameras, citizen scientists have documented wildlife at every single one of them! Your classifications really do give us insight into how our animals look at and make use of the landscape.

Match 'em up with the first trio of images in these posts - there really are wildlife using these areas! And you all as citizen scientists are the ones finding them!

Another concern was wanting to make sure we had hard evidence to back up any off-grid placements before we started moving cameras around. Being able to back up deviations from our original design with data, and justify them to reviewers when we publish the results from this study is important. We are discussing other subexperiments we could design to test whether there is an effect of camera height, orientation and/or placement on the animals we get pictures of. Stay tuned!

Of course, we do appreciate firsthand that pictures of cars driving by (or of nothing because a car trigged the camera and was going so fast it got out of the frame before the camera took a picture!) are NOT why you are looking through project images! Nobody, not even the most dedicated scientists, likes going through thousands of pictures of vehicles. Like you, there have been days when I have given up classifying in frustration after clicking through car after car after car, without even a deer to break up the monotony. We are working on an algorithm that will hopefully automatically classify and remove most of the images like this from the subject sets in the future, so that you can spend your time looking at wildlife instead of cars! Check back this weekend for a post about machine learning and our plans to integrate it into this project. In the meantime, we greatly appreciate you persevering through these less-than-ideal images. There's no need to waste your time hashtagging them, just click 'Human or Vehicle' and move on - hopefully to a picture of a fox, raccoon or pheasant!