×

Field biology shows that GoPro cameras aren’t just for snowboarding daredevils

  • A beaver dam and pond in the Hubbard Brook Experimental Forest in Woodstock, N.H., in June 2015. This is one of the areas where KSU Prof. Denise Burchsted and undergradaute Mike McGuinness testing using GoPro cameras to gather data in this murky, difficult environment. Denise Burchsted—Keene State University

  • A new beaver dam and pond in the Hubbard Brook Experimental Forest in Woodstock as seen in June 2015. This is one of the areas Denise Burchsted and Mike McGuinness observe with GoPro cameras. Denise Burchsted / Keene State University

  • KSU student Mike McGuinness operates an underwater GoPro camera affixed below white buoy in 2015. Courtesy of Denise Burchsted / Keene State University



Monitor staff
Tuesday, May 09, 2017

An important skill for scientists is knowing where to spend your funding, which explains why Keene State College’s Denise Burchsted sounded proud when she described her protocol for studying beaver ponds: “Rather than investing in winches to pull undergraduate students out of the muck, I invested in a few GoPro cameras.”

Don’t worry, KSC parents: This assistant professor of environmental studies isn’t abandoning biology majors to the Cheshire County mire. Rather, Burchsted is developing a new – and cheap – way to understand the difficult-to-investigate ponds that are created by beaver dams, which are a focus of her research partly because they haven’t been analyzed as much as you might think.

“It’s hard to study beaver ponds. Boating is usually difficult, because the ponds are too shallow and full of dead trees (“gunk holes,” as my colleague says). Wading is also difficult, because the bottom is incredibly mucky,” is how Burchsted described the situation.

It’s well-known that beaver ponds are important for the New Hampshire’s ecosystem, since they’re rich in nutrients and organic material, and are home to a variety of invertebrates, fish and birds, but even a very basic fact like how many fish live in them is hard to obtain, she said.

The easiest method to determine fish populations is electroshocking, in which fish in part of a stream or lake are stunned by an electric charge and float to the surface for easy counting. That doesn’t work in a cluttered beaver pond, because the stunned fish get caught on all the woody debris. As a result, most beaver-pond population data depends on traps, which can be inaccurate.

Puzzling over this problem, Burchsted got an idea from an unexpected source. “My then-11-year-old son said Mom, you should try a GoPro camera.”

She looked at these small, rugged cameras, best known for their use in making action-filled “selfie” films while skiing, kayaking or mountain climbing, and liked the idea. She knew that one of her students, Mike McGuinness, liked fishing so she thought he’d be just the person to splash through the muck and water, and experiment with ways of placing GoPro cameras in the wild.

“He spent a summer out in the field and looked for fish in beaver ponds with a GoPro camera. It was so much more successful than I would have imagined ... There were more than twice as many fish we caught on camera as compared to traps. Not just numbers, but species. With a trap we would get one species, with a camera we’d get three or four,” she said. “We had to develop new methods to count them, there were so many.”

They collected underwater video more than 100 times at sites along 12 rivers and streams in New Hampshire, with an “in-channel” beaver pond having by far the highest number and diversity of fish.

“Some of our favorite videos show fish swimming in and out of a minnow trap, as though there were almost no barrier,” she said – a reflection of why trap data are suspect.

Some videos taken by the team are linked from the online version of this article.

Gathering good data is important to a research scientist, of course, but gathering good data without spending much grant money is even better.

“This is really low-cost. You don’ t need any fancy equipment. Just an undergrad and myself can generate data,” Burchsted said of using GoPro cameras. “It’s almost to the level of citizen science.”

Interestingly, Burchsted, whose pre-academic life as an engineer involved helping remove human dams to improve river systems, admits she isn’t all that interested in fish and frogs and aquatic bugs: “I am more interested in the nonliving parts” of the environment. But you can’t understand the nonliving parts without understanding what the living parts do to them, hence this research.

What she really wants to accomplish, she said, is to understand how beaver ponds work so we can re-create their benefits in urban environments without having to import beavers, which will promptly fill in all our culverts.

“Beaver ponds are a part of the river system that was here before Europeans, and we want to know what functions they provide, how they contribute to the nutrient cycling,” she said.

“We’re trying to get a handle on what types of functions they are providing – what can we do to provide some of these functions that we’ve lost, that the beavers would be providing for us. (We want) kind of a recipe for what we can do to create more of these places, to support all of the life stages of in-stream habitat, not just fish.”

That’s why she thinks one of the most interesting videos collected by the GoPro “comes from urban Keene, where we see a relatively high density of fish ... and a high species richness. It seems possible that these urban locations would provide habitat similar to a beaver pond.”

Pretty cool. But this is science, so it’s not that simple.

Burchsted said the GoPro project was a “proof of concept” and now she’s developing methods and processes to establish its bona fides. She needs to show that the results weren’t a lucky accident but a better reflection of what is going on in these ponds, which provides other researchers confidence in the results.

“We have some work to do from here, applying it and comparing it with standard methods,” she said.

Still, if all goes well, the toolkit for field biologists might one day include studying GoPro use as much as the work of Nobel Prize-winning scientist Konrad Lorenz.

(David Brooks can be reached at 369-3313 or dbrooks@cmonitor.com or on Twitter @GraniteGeek.)