Forensic Fishing: Using eDNA to Track Fish Populations

Mike Spear heads onto Bearskin Lake looking  for environmental DNA — or eDNA—the genetic material fish shed as they move around the lake.  Photo: S. Widell

by Sydney Widell – At the start of every fishing season, the Wisconsin Department of Natural Resources embarks on the arduous task of setting catch limits on the state’s popular game fish like walleye.

In order to estimate how many fish anglers may safely remove from a lake while still keeping the population healthy, the WDNR first needs to estimate how many fish are in the lake to begin with and how those populations are changing. Not an easy task, considering that there can be hundreds of thousands of fish in a hundred-acre lake. And more than 1,000 of the state’s 15,074 lakes fit that bill.

Enter Center for Limnology graduate student Mike Spear, who may be on the brink of developing a radical new way to estimate fish populations by decoding their DNA.

Mike is recording eDNA data on lakes that the DNR has already surveyed this year, in order to compare his population estimate to theirs. Photo: S. Widell

Mike is based in Madison, but I got the chance to catch up with him in the field on Bearskin Lake when he came up north to sample last week. He brought Dane McKittrick, one of his undergraduate assistants, to help. As we surveyed, Mike told me more about his project.

As fish swim through the lake, they constantly shed genetic material in the form of skin, eggs and waste, Mike explained. The amount of environmental DNA — or eDNA — in the water might give researchers like Mike clues about the size of the fish populations it came from.

“These fish are literally broadcasting their presence in DNA,” Mike said. “What we want to do is see if we can use environmental DNA to get population estimate information in a much quicker, less disruptive and less costly way.”

Right now, Mike is specifically looking at walleye — a species that the WDNR is mandated by law to make population estimates of annually. The WDNR collects data on 15 different lakes each year, and Mike is surveying its 2018 sample sites. That way, he’ll be able to compare his findings to theirs.

Dane McKittrick, one of Mike’s lab assistants, takes a water sample. They’ll collect water from nine sites across this lake, sampling at a range of depths and habitats. Photo: S. Widell

Traditionally, the WDNR — as well as the Great Lakes Indian Fish and Wildlife Commission — have used a mark and recapture method to count walleye before the fishing season opens in the spring.

In the spring, researchers capture and tag walleye on a lake using nets. Later on, they return to the lake, and using a method called electrofishing to stun fish, quickly net them and check for tags before safely releasing them back to the lake.

Based on sex ratios, the size of the lake and the number of tagged walleye they caught a second time, they can extrapolate the size of the entire population.

But, for the WDNR, surveying the 15 lakes takes weeks and requires a lot of personnel.

Could fisheries managers one day estimate fish populations in Bearskin Lake just by taking a few water samples – instead of weeks of intensive netting? Photo: S. Widell

If Mike can establish a relationship between the amount of walleye eDNA in a lake and the population estimates made by the WDNR, he could potentially model entire populations based entirely on eDNA.

“The idea behind that is we have all these walleye swimming around in a lake, and it takes a lot of time and effort to pull those walleye out of the lake and count them,” Mike said.  “But we can, just by taking a water sample, find and count the number of walleye DNA molecules that are floating around in the water by identifying the walleye DNA, counting the molecules, and creating what is called a copy number —  the concentration of walleye DNA per liter of water.”

As he explains, Mike steers us around the lake from point to point, stopping every so often in what feel like random locations to sample. But they’re not random — he selected them based on their diverse depths and habitat features.

“We do nine samples across the lake and we try to get a mix of the nearshore sites and offshore sights,” Mike said. “And among the near shore, we try to get a mix of different habitat types, so we’ll do one among the weeds, and we’ll do one around some coarse woody debris.”

Dane is the one collecting the water samples. When Mike stops the boat, Dane pulls on a pair of latex gloves — the team takes contamination risks seriously — and fills a jar with what Mike describes as “a glug” of lake water. Then, he places it in a cooler that Mike will take to UW-Madison for analysis later on.

Dane stores the sample in a cooler to slow degradation of the DNA samples. The samples will be frozen and tested over the course of the next year. Photo: S. Widell

In Madison, Mike will spend the fall extracting and measuring walleye DNA from the samples Dane is taking right now. They will be looking for a segment of genetic code unique to walleye, or a “walleye signature,” as Mike likes to say.

This is the second year Mike has done this survey. When he ran it last year, he told me his results looked a lot like the WDNR’s.

“The relationship was strong enough that we’re doing it again this year to see if it holds up year to year,” Mike said. “It could be a tool going forward that the WDNR or GLIFWC might want to adopt.”