Can contemporary evolution explain the many enigmas in recent dynamics of Norwegian spring-spawning herring?

Informations

Funding country

Norway

Acronym

-

URL

-

Start date

1/1/2015

End date

12/31/2018

Budget

914,259 EUR

Fundings

Abstract

Norwegian spring-spawning herring is one of the largest pelagic fish stocks in the world, and likely also one of the best studied and documented ones, with time series data on catches back to 1899, on population dynamics back to 1904 and biological data on individuals starting from 1935. Nevertheless, the dynamics of the population have remained enigmatic and difficult to predict, and consequently the stock assessments have overestimated the population size year after year. Additionally, no evolutionary life-history changes in age or size at first spawning in response to substantial fishing pressure have been observed, as theory would predict and in contrast to many demersal fish species such as Atlantic cod. This has raised the question what really drives dynamics in pelagic stocks such as Norwegian spring-spawning herring and if their response to the selection pressure caused by harvesting could be different to what has been observed in demersal stocks. In this project we have utilized the massive amount of data on Norwegian spring-spawning herring, other species and the ecosystem collected over the years, and applied life history theory and a combination of statistical and simulation modelling, to examine how these observed patterns are connected and shed some light on the environmental, anthropogenic and evolutionary drivers of population dynamics. As initial approach to these questions, we used broad comparative analyses of recruitment dynamics in Northeast Atlantic fish stocks and found in three articles (one published, one provisionally accepted, one in preparation) that 1) density-dependent recruitment is the dominating internal driver of population dynamics, 2) showed how variation in mortality at early life stages affects the stock-recruitment relationship, and 3) determined common trends and underlying environmental drivers of recruitment variability. In an additional study published we also analysed that reductions in fishing mortality were key to the recoveries observed in many Northeast Atlantic fish stocks in recent years but recruitment success was also a crucial factor. To present and discuss these findings in the international scientific community, we organized and convened a theme session on recruitment dynamics at the ICES Annual Science Conference 2017. Secondly, we developed a simulation model to explore how behaviour-selective fishing that targets, e.g., schooling or actively hunting fish affects life-history traits such as growth, natural mortality and maturation age, resulting in one published article and a presentation at an international conference. These studies laid the groundwork for more detailed investigations of dynamics in NSS herring that investigate changes in growth and reproductive investment over time, using the existing time series of scale increments, length and weight, and gonad weight. Three papers presenting the results are currently in preparation. Furthermore, in a current analysis we are exploring how natural mortality in NSS herring has potentially changed over time and how slipping or net bursts in the fishery may cause additional mortality, affecting population dynamics and stock estimates. Because available time series of tagging data were found to be too inconsistent for comprehensive statistical analysis, we used existing information to form hypotheses on changes in natural mortality over time and focus on how such unaccounted mortality causes bias and error in stock assessment estimates. Part of the project was conducted at the University of Glasgow, UK, and the University of California Santa Cruz, USA, by the PhD student and postdoc, respectively, between 2017 and 2018, resulting in an experimental study on behavioural selection, a novel method to estimate stock productivity from life-history parameters, and the development of a statistical approach to predict nonlinear dynamics in recruitment. By January 2019, the PhD position within the project has resulted in a successfully defended PhD thesis. All the remaining project-related studies and corresponding scientific articles are well advanced or submitted and expected to be finalized during 2019. Furthermore, the insights gained and hypotheses formed within the project will be further investigated by the involved researchers in their future work, notably as part of two PhD positions at the University of Bergen. The project and its progress has also been presented to the broader public, foremost through social media and the project website conevolher.imr.no.