Amid Western Alaska salmon crisis, data-driven strategies could reduce chinook bycatch

a biologist
Sabrina Garcia, a marine research biologist with the Alaska Department of Fish and Game, prepares to release a chinook salmon bearing a pop-up satellite archival tag (PSAT) outside of Craig, AK as part of training for her PhD research at the University of Alaska Fairbanks in May 2022. (From UAF)

In the debate over what is driving the Western Alaska chinook, or king, salmon crisis, the commercial trawl industry has faced no shortage of criticism over the issue of bycatch. The vast majority of chinook bycatch takes place in the Alaska pollock fishery, the second-largest fishery in the world.

Each year, thousands of chinook that would otherwise make their way to the Yukon and Kuskokwim rivers are intercepted at sea. In 2020, more than half of the estimated roughly 32,000 chinook caught by the Bering Sea and Aleutian Islands pollock trawl fisheries were from Western Alaska stocks, according to the National Oceanic and Atmospheric Administration. This was also the year the Yukon River was completely closed to subsistence king salmon fishing for the first time ever, and it hasn’t reopened since.

Chinook salmon are federally managed as a prohibited species catch, or PSC. While the pollock fleet has remained within the bounds of the cap currently in place for chinook bycatch, researchers are looking at ways to further reduce the number of the fish that get scooped up.

Sabrina Garcia, an Alaska Department of Fish and Game biologist, is in the middle of a five-year tagging and modeling project as part of her Ph.D. studies at the industry-funded University of Alaska Fairbanks Pollock Conservation Cooperative Research Center. Garcia hopes that the project will generate predictive maps for chinook bycatch that could sharpen the way the Alaska pollock fishery is regulated. To do that, the computer models have to be tested against real data.

“There is a chance that we try to create these models and we don’t find that winning combination of variables that accurately predict chinook salmon presence,” Garcia said. “These models are being used in other fisheries and they work for things like turtles and whales. Can we apply that to chinook salmon?”

Currently, pollock trawlers operate under rolling hotspot closures that change based on real-time chinook bycatch data that vessels provide to a private monitoring group called Sea State. While the system has helped reduce chinook bycatch, Garcia said that it lacks the fine-scale data about the depth where the salmon were caught that could help researchers understand their ocean habitat. She hopes the predictive maps her team is working to develop can offer trawl captains another tool to avoid bycatch.

“Ultimately, this is going to be a tool for them. And if they don’t have input along the way, and they don’t have buy-in along the way, then it’s not going to work,” Garcia said.

But to create the necessary models, the researchers need reliable data from what are known as pop-up satellite archival tags, or PSATs. At $4,000 a piece, these are small devices programmed to detach from mature chinook salmon after nine months at sea, float to the surface, and transmit a whole lot of information about what those fish have done. Garcia hopes that this data will yield brand new insights about chinook migration patterns, especially stocks from Western Alaska.

“They get attached to the chinook on a little backpack. And what those tags do is they are recording depth, temperature, and light level,” Garcia said.

Combined, these three metrics track the movements of the tagged fish in the ocean.

“We have temperature at depth, and we know the rate at which light gets filtered through the water column,” Garcia said. “These amazing models can use that information to basically geo-locate that information.”

Garcia’s project advisor, Andy Seitz, has already deployed 35 of the tags in the Bering Sea, and Garcia is gearing up to attach 19 more of them next year. It isn’t easy. She said that catching chinook using rod and reel gear on the open ocean is like finding a needle in a haystack. Even when it does work, sometimes it’s in vain.

“Some of them can die, some of them can get eaten by a salmon shark, some of them can get eaten by a whale. Some of them can die just because they didn’t have enough food in their stomachs,” Garcia said.

As she juggles her studies, research and full-time employment with the state, Garcia said that she is more eager than ever to dive into the chinook data that will soon begin pouring in. Whether her team will ultimately be successful incorporating these data into predictive maps to be combined with the system already in place remains to be seen.

“I’m hoping beyond hope that this is something that does help industry reduce bycatch,” Garcia said.

But this tagging and modeling project is just the beginning. Garcia said that she’s already making plans to go on to research what happens at sea to chum salmon as well.

Previous articleSitka denies petition to put cruise passenger limits to voters
Next articleTalk of Alaska: The rise in book challenges