Dynamics of spawning migrations in Northeast Arctic cod

Abstract

The Northeast Arctic (NA) cod is known to undertake long southbound spawning migrations from their feeding grounds in the Barents Sea, to various spawning grounds along the Norwegian coast. Hence it’s native Norwegian name “skrei”, meaning to move or to travel. The spawned eggs and larvae subsequently drift northwards along with the prevailing currents, eventually reaching the Barents Sea as juveniles. From commercial fisheries statistics we see that these spawning grounds once spanned across nearly 2000 km of the west coast of Norway, from Finnmark in the north to Vest-Agder at the very south. Today the latitudinal range of the spawning grounds is significantly reduced, with only a marginal part of the stock spawning as far south as Møre, some 1500 km from the Barents Sea. Coarse estimates from the Institute of Marine Research suggest that around 90 - 95 % of the spawning stock spawn from Lofoten and northwards, utilizing only a third of their original expanse. Also, recent scientific surveys have found that during the last decade, the NA cod have to a great extent stopped spawning at their major traditional grounds in Vestfjorden, Lofoten, and have relocated to more northbound areas. Historically, scientists have shown little interest in the changes in spawning grounds, but due to the major impact this recent shift has had on local fishermen and landing ports, the topic currently attracts noticeable interest from fisheries scientists as well as climate scientists. The NA cod’s previous widespread spawning distribution, together with its historical and contemporary northbound shifts has spurred two major research questions addressed in this thesis. First, why does spawning take place over such a vast geographical area, implying that certain parts of the population undergo spawning migrations several thousands of kilometres further than their conspecifics? Secondly, what has caused the northbound shift in spawning grounds? From general evolutionary theory, we would expect that individuals undertaking longer spawning migrations, thus leaving less time to forage in the Barents Sea, as well as increasing their energetic cost of migration, would achieve a comparable return benefit. If not, such a life history strategy could simply not hold through the course of natural selection. The benefit need not target the spawning individual directly, but can also be mediated through increased fitness to their progeny, essentially through increased survival probability. Using a simplistic assumption that fitness benefit increases linearly with migration distance, we developed a model simulating a population of individuals which finds optimal solutions to the trade-off between growth and reproduction, depending on physiological condition and ecological constraints. Overall, the model predict that larger individuals and individuals in better condition gain higher fitness benefit from longer spawning migrations compared to smaller and less fit individuals. These findings are partly due to a nonlinear relationship between hydrodynamic friction and individual size, meaning that relatively, larger fish spend less energy on swimming compared to smaller individuals. More interestingly, when simulating historical fishing pressure at the spawning grounds, there is selection for large late-maturing fish and longer migrations, whilst a contemporary trawl fishery, typically located at the feeding areas in the Barents Sea, select for small and early-maturing fish with shorter optimal migration distances. The latter case is consistent with observational studies, and indicates that fisheries’ induced evolution have not only lowered the maturation age of NA cod, but may also be causing the northbound shift in spawning ground distribution. To test the validity of our initial assumption that southerly spawning grounds are in fact associated with higher fitness benefits, we employed different general circulation models to track virtual fish eggs and larvae released at various spawning grounds along the Norwegian coast. From their drift trajectories towards the Barents Sea, we found that eggs and larvae released from more southerly spawning grounds experienced higher average temperature exposures, generally thought to promote faster growth and consequently reduce mortality in early life stages. However, the southernmost spawning grounds generally also experienced more retention in local fjord systems. In addition, seasonal and inter-annual variation in drift trajectories, as well as overall temperature exposure, growth and survival was evident, indicating that climatic conditions may also play a role for offspring success. Overall, the latitudinal effect on larval temperature exposure was significantly stronger than the climatic variability. Finally, by utilizing empirical data from commercial catch statistics dating back to 1866, our initial theory, that shifts in spawning grounds are caused by a size-selective industrial trawl fishery in the Barents Sea, was tested against alternative explanatory factors such as density dependence and climate change. In total, 104 years of landing data were compiled for the entire Norwegian coast, revealing large fluctuations in spawning ground distributions, but also showing trends towards more northbound spawning after the 1920s. Climatic variation was found only partially to explain the variation, whilst rapidly increasing landings from the trawl fishery in the Barents Sea starting around 1923, clearly coincided with the northbound shifts in spawning grounds.