Anders Frugård Opdal

THE EVOLUTION OF SPAWNING MIGRATIONS: STATE DEPENDENCE AND FISHING‐INDUCED CHANGES

Individuals migrate to exploit heterogeneities between spatially separated environments to modulate growth, survival, or reproduction. We devised a bioenergetics model to investigate the evolution of migration distance and its dependence on individual states. Atlantic cod Gadus morhua ranges from sedentary populations to stocks that migrate several thousand kilometers annually. We focused on the Northeast Arctic cod stock, which migrates south to spawn. A linear relationship between migration distance and the expected survival of offspring was assumed, here understood as the prospects for future survival and development that a fertilized egg faces at a particular spawning location. Reasons for why it may increase southward include warmer water that increases development rates, and thereby survival, along the pelagic drift trajectory. In the model, ingested energy can either be allocated to growth or stored for migration and reproduction. When migrating, individuals forgo foraging opportunities and expend energy. Optimal energy allocation and migration strategies were found using state-dependent optimization, with body length, age, condition, and current food availability as individual states. For both a historical and contemporary fishing regime we modeled two behaviors: (1) homing cod returning to the same spawning site each year and (2) roaming cod with no such constraints. The model predicted distinct regions of locally high spawning stock biomass. Large individuals in good condition migrated farthest, and these also tended to mature later in life. The roaming cod spread farther south as they grew larger and older. Homing cod did not have this freedom, and spawning was generally concentrated along a narrower stretch of the coastline. Under contemporary fishing, individuals matured earlier at a smaller size, had shorter migrations, spawned over a contracted geographical range, and tended to be in poorer condition. The effects were most pronounced for the homing behavior.

Long‐term change in a behavioural trait: truncated spawning distribution and demography in Northeast Arctic cod

Harvesting may be a potent driver of demographic change and contemporary evolution, which both may have great impacts on animal populations. Research has focused on changes in phenotypic traits that are easily quantifiable and for which time series exist, such as size, age, sex, or gonad size, whereas potential changes in behavioural traits have been under-studied. Here, we analyse potential drivers of long-term changes in a behavioural trait for the Northeast Arctic stock of Atlantic cod Gadus morhua, namely choice of spawning location. For 104 years (1866–1969), commercial catches were recorded annually and reported by county along the Norwegian coast. During this time period, spawning ground distribution has fluctuated with a trend towards more northerly spawning. Spawning location is analysed against a suite of explanatory factors including climate, fishing pressure, density dependence, and demography. We find that demography (age or age at maturation) had the highest explanatory power for variation in spawning location, while climate had a limited effect below statistical significance. As to potential mechanisms, some effects of climate may act through demography, and explanatory variables for demography may also have absorbed direct evolutionary change in migration distance for which proxies were unavailable. Despite these caveats, we argue that fishing mortality, either through demographic or evolutionary change, has served as an effective driver for changing spawning locations in cod, and that additional explanatory factors related to climate add no significant information.

Response: Demography affects spawning location in Northeast Arctic cod, but what affects demography?

Uni Research and Hjort Centre for Marine Ecosystem Dynamics, P.O.Box 7810, 5020 Bergen, NorwayIn a letter to the editor, Sundby (2015) expresses con-cerns about our study of changing spawning locationsof the Northeast Arctic (NEA) stock of Atlantic codover the period 1866–1969, where we identified statisti-cally significant effects of the stock’s demographywhereas various climate indices all fell below statisticalsignificance. Our conclusion on the role ascribed to cli-mate disagrees with that of Sundby & Nakken (2008),which, based on a subset of our data and without con-sidering demography, concluded that spawning wasshifted northwards in warm periods.Sundby’s (2015) criticism boils down to four pointsthat we will address in turn.A: ‘the calculated spawning migration distance (...)is based on incorrect assumptions’ (quote fromabstract, covered by points 1–3 in Sundby’s letter)In our analysis, we never did calculate spawningmigration distance because the data did not allow it.From the data, a time-series of regional statistics of thecod spawner fishery, we could only quantify spawninglocation. We thus only know the endpoint of the spawn-ing migration, and as there are no data on where in theBarents Sea migration began we could not, and did not,calculate migration distance. It is unclear to us how thismisreading of our paper originated. In fact, we explic-itly raise this issue, for example in this quote (Opdal &Jorgensen, 2015, page 1527):It is also known that younger cod tend to be dis-tributed in colder water further north and east inthe Barents Sea than older, larger cod (Ottersenet al., 1998). This has the implication that one can-not infer from spawning location how far an indi-vidual has migrated to get there.Instead of using metrics of latitude and longitude, wequantified spawning location as distance from a centralpoint in the Barents Sea. This is only a technical issue ofpresentation and does not make assumptions or infer-ences about migration distance.B. An alternative hypothesis is that feedingdistribution in the Barents Sea explains spawninglocation (point 3 in Sundby’s letter)Sundby (2015) proposes the hypothesis ‘that the spawn-ing migration distance is constant, since the distribu-tions during the feeding in the Barents Sea and at thespawning areas are fluctuating with the similar ampli-tude (i.e. 250-300 km) on decadal scale’ (see fig. 1 inSundby, 2015). He also notes that ‘data is not availableto test this’. However, Sundby (2015) also refers to sev-eral studies showing that feeding cod move furthernorth and east in the Barents Sea during warm years(e.g. Kjesbu et al., 2014). With temperature-dependentfeeding location and constant migration distance, a pre-diction from his hypothesis is thus that spawning loca-tion would shift north in warm years, which is exactlythe effect that our statistical analysis finds no supportfor.C. We are using irrelevant climate indices (point 1in Sundby’s letter)Sundby stresses that the North Atlantic Oscillation(NAO) is an atmospheric pressure index and not a mea-sure of water temperature, which is what likely wouldhave the strongest effect on fish. We are of courseaware of this, and included NAO only because directobservations of sea temperature (the Kola section) wereinitiated in 1900 and therefore not available for the first34 years of our data. Neither NAO nor Kola sea tem-perature significantly explained variations in spawninglocation in our study. The use of NAO is not completelyunwarranted, however, as NAO has been found to cor-relate with oceanography and ecology of cod in theBarents Sea (Ottersen & Stenseth, 2001). More gener-ally, pressure indices describe large-scale physical driv-ers, with influences on the dynamics of both terrestrialand marine ecosystems (Stenseth et al., 2002).