“Collapse, tipping points, and spatial demographic structure arising from the adopted migrant life history”

Posted on

Luke A. Rogers, Anne K. Salomon, Brendan Connors, and Martin Krkošek (July 2018)

Read the Article

Socially learned behavior may influence fish dynamics like collapse and tipping points

“Herring have lost their big Chiefs and no longer know where to spawn.”

On the Central Coast of British Columbia, an Indigenous Hereditary Chief and herring fisherman offered his observations that the number of large Pacific herring was declining. His knowledge became the focus of our research.

It’s difficult to study how open ocean fish species like herring and cod learn and maintain their migration routes to find areas to feed and spawn year after year. But according to new research, behaviour that young fish learn from nearby older fish—called social learning—may be critically important to maintain the health of fish stocks.

A theoretical model that mirrors the “Big Chiefs Hypothesis” proposed by the Indigenous Hereditary Chief was developed by Luke Rogers and Martin Krkosek of the Department of Ecology & Evolutionary Biology at University of Toronto and colleagues from Simon Fraser University and Fisheries and Oceans Canada. The model helped the researchers ask, if small juvenile fish learn migratory routes and the location of spawning habitats by following larger, more experienced adult fish, what is the effect on the survival of fish populations?

“This theory could help explain why some fish populations that collapse, like some Atlantic cod populations in Newfoundland, take an unexpectedly long time to recover,” says co-author Brendan Connors. “It could also explain why good habitats are not always quickly recolonized following collapse, as has been the case for some Pacific herring populations in British Columbia.”

The researchers found that when the abundance of local fish declined, juvenile fish failed to encounter and learn from adults. This lack of learned experience eventually caused smaller fish populations to go extinct. And once a local population was lost, only a chance event could reestablish it, although when and how often this might occur is still unclear.

“Our results suggest that local fish populations could be vulnerable to local collapse if, for example, some local populations faced much more fishing pressure than others,” says Rogers, lead author of the study.

Commercial fisheries are typically managed separately for different species, with a focus on the regional abundance of fish. This could—in principle—lead to a situation in which local populations are depleted by a fishery, one after the next after the next, with little recognition of these local declines at the larger scale of regional management.

“It’s very scary now with the rate of exploitation of some species where they are fished to near extinction and how that will affect other species and us,” says Mike Reid, Aquatics Manager for the Heiltsuk Integrated Resource Management Department (HIRMD) and research Partner on the NSERC grant that funded the study.

“Commercial herring fisheries in Haida Gwaii have been closed for over a decade now. The depletion of larger, older fish could explain in part the failure of our herring populations to rebuild,” says Russ Jones, retired Manager of Marine Planning for the Council of the Haida Nation.

While the model reaffirmed the Big Chiefs Hypothesis, the research team wants to go further. Similar to a flight recorder, unique chemical tracers of where fish were born are left behind in the ear bones of fish. The research team hopes to use these ear bones—plus knowledge of local population age and size structure—to test whether social learning by herring occurs in wild fish populations.

“When you look at the definition of science—‘the intellectual and practical activity encompassing the systematic study of the structure and behaviour of the physical and natural world through observation and experiment’—this is exactly what we do over millennia. We are taught to observe our environment at a very young age through harvesting and fishing practises and this knowledge is passed down,” says Reid. “This research explains what we’ve said for a long time was truth for herring.”

“The work was both catalyzed and enriched by the breadth of traditional knowledge that our Heiltsuk and Haida First Nations research partners shared as collaborators,” says senior coauthor Martin Krkosek.

“It’s commonly assumed that Western science is the primary tool to initiate research questions, generate hypotheses, collect data and produce knowledge. As such, western science is routinely used to define, claim and arbitrate ‘truth’,” says co-author Anne Salomon.

“Our discovery, generated through a research partnership with Indigenous knowledge holders, challenges that assumption,” says Salomon. “It is a beautiful example of how a local observation, grounded in place-based knowledge accumulated through generations, generated a hypothesis that led us collectively to a ‘truth’ that holds legitimacy among all of us involved in the research partnership. This is a way forward in shared, evidence-based decision making in Canadian fisheries.”

This research was funded by a Natural Sciences and Engineering Research Council of Canada (NSERC) Strategic Project Grant, the Hakai Institute, and a Canada Research Chair.


The roles of dispersal and recruitment have long been a focal point in ecology and conservation. The adopted migrant hypothesis proposes a life history in which social learning transmits migratory knowledge between generations of iteroparous fish. Specifically, juveniles disperse from the parental spawning site, encounter and recruit to a local adult population, and learn migration routes between spawning and foraging habitats by following older, experienced fish. Although the adopted migrant life history may apply to many species of pelagic marine fishes, there is scant theoretical or empirical work on the consequent population dynamics. We developed and analyzed a mathematical model of this life history in which the recruitment of juveniles depends on the relative abundance of the local populations and recruitment overlap, which measures the ease with which juveniles are recruited by a non-parental population. We demonstrate that the adopted migrant life history can maintain spatial demographic structure among local populations, that it can also predispose local populations to collapse when a tipping point is crossed, and that recovery after collapse is impaired by reduced recruitment at small local population sizes.