Article body copy
A baby salmon is, often, the product of its parents’ final burst of life. The two adult fish, having fought their way upriver—skirting predators and navigating sometimes thousands of kilometers inland—end up hovering together, spewing eggs and sperm over the same rocky streambed where they were born. Likewise, their offspring will migrate out to sea, spending several years gobbling up plankton and fish before returning to that stream to start the cycle anew.
Except, occasionally, they don’t. Sometimes, salmon stray on their way home, venturing into unknown waters. Each year, some percentage of migrating salmon always stray, but scientists don’t know why that percentage can vary so much—some years there are many strays and other years there are only a few. Now, some researchers have an idea why, and their work has helped foster a new scientific discipline.
It all began in 2013 when Peter Westley, now a fisheries ecologist at the University of Alaska Fairbanks, discovered something counterintuitive. Westley was parked in front of a computer crunching salmon straying rates and noticed that in years when salmon are abundant, fewer stray than in years when they are scarce. To Westley, that didn’t make sense. “If there’s competition on the spawning grounds, you would think some fish would go elsewhere. Yet the numbers said the opposite.”
Around the same time, Andrew Berdahl, now an ecologist at the University of Washington, was studying the movements of the golden shiner, a minnow that prefers darkness to light. Berdahl had placed groups of golden shiner in a tank, shone a light, and moved it randomly. He found that individual fish could not navigate to the dark areas on their own. But when they were in a big enough group, they could. When one fish found a dark patch, Berdahl says, it would slow down, prompting the whole group to turn toward it.
When Westley found Berdahl’s write-up, he had an epiphany: maybe the straying salmon just didn’t have enough colleagues around to make an informed decision about which way to go. “The pattern in [Berdahl’s] data mirrored what I was seeing,” Westley says. “I thought, maybe individual salmon are more likely to get lost if they’re in smaller groups because there’s not enough collective wisdom. It was an aha moment and [it] led me to reach out to Andrew.”
The two began a partnership that, in the years since, has resulted in a new hypothesis on how groups of fish make decisions, and the development of a new field of science called collective movement ecology.
“We saw this disconnect between classic movement ecology, that looks at animal movement in time and space but that’s largely done on an individual level, and social collective behavior, that looks at the finer-scale aspects of individual social interaction within fish schools, insect swarms, animal herds,” says Westley.
In a series of papers, the two have laid out their case, arguing that salmon in an upriver run make decisions collectively.
The simplest piece of evidence is an expanded version of the discovery Westley had identified several years earlier: when there are more salmon, the number of strays decreases, and vice versa. But other evidence is more involved. Studies tracking salmon movements have shown that salmon gather at tributaries before turning into them, but could not explain why. Westley and Berdahl suggest the salmon may be sampling the conflicting odors where the water forks, then gathering on the side they guess to be the correct stream. Eventually, they come to a majority decision and head that way. “The idea is that they’re essentially voting,” says Berdahl.
In a paper, the two show how their hypothesis offers a better explanation for a well-known phenomenon. When salmon enter their native spawning grounds, they do so in pulses: one day 200 might enter, then none for several days, then 500, and so on. “Folk wisdom attributes these pulses to changes in water temperature, flow rates, tides, the moon, whatever,” Berdahl says. But when scientists have tried to measure such influences, the results have been contradictory. Berdahl and Westley think the fish might be waiting for a social cue: as one fish feels ready to spawn, it moves forward. That triggers the next, resulting in a cascade of movement.
To test their idea, they created a simple computer model that allowed simulated fish to enter a creek either using social cues—following the fish in front of it—or by making independent choices. In the simulation, the social model produced outputs that nearly perfectly matched what was observed across 30 years of sockeye salmon data in Hansen Creek in Alaska’s Bristol Bay region. “It produced the same pattern that was found in nature,” Westley says. “Working on it reminded me how beautiful science can be in its simplicity and elegance.”
Nolan Bett, who as a postdoctoral researcher at the University of British Columbia helped show how salmon follow a hierarchy of olfactory navigation cues when migrating upriver, says Westley and Berdahl’s collective migration theory makes intuitive sense. “Anybody who’s spent time standing at the side of a river looking at these salmon runs notices that they come up in these groups,” he says. “And so it seems reasonable to think there could be some kind of interaction between them and that it benefits them to be in large groups when they’re trying to do something that is very hard to achieve, which is migrating thousands of kilometers to a specific site,” Bett says, adding: “The data that they have so far is, I would say, preliminary, but it certainly points in the right direction.”
Westley uses the tragedy of the northern cod to illustrate the importance of understanding social movement. In 1992, east coast cod populations collapsed practically overnight. Fisheries managers were unaware of just how low cod numbers had become because they relied on catch rates as indicators of fish abundance. But cod travel in schools, and their declining numbers caused them to school more tightly than before. Fishermen were still catching plenty right up until the end. “It’s a prime example of why we need to understand these collective movement dynamics, ” Westley says.