CoDINA - Cod: DIet and food web dyNAmics

What determines the menu of the Barents Sea cod - a model study on the cod's diet and the Barents Sea food web The project CoDINA-Cod: DIet and food web dyNAmics is a Norwegian-British-Russian joint project led by the Institute of Marine Research. CoDINA addresses the connection between the Barents Sea cod and its main prey. Capelin, smaller cod (cannibalism), other species of fish and various crustaceans are the most important elements in the menu. A central question is how this interaction is affected by the size of the cod stock, the presence of many large cod and the general state of the ecosystem. The aim of CoDINA is to increase the understanding of the dynamics of the food chain in the free water masses of sub-Arctic ecosystems. More specifically, we study predator-prey interactions in the Barents Sea. Our starting point is especially the variability in cod diet compositing in space and time. We examine how this is related to variation in the environment, including climate, and characteristics of the stocks of cod itself and their main prey. We model multi-species interactions with various types of statistical (data-driven) and more analytical models. The quality of results from statistical analyses and models depends on the data being available. This is especially the case when studying phenomena that vary at different scales in both space and time. Therefore, preparation and quality control of data within the Norwegian-Russian stomach database was a central task in the first phase of CoDINA. Information from more than 269,000 stomachs from 1984-2016 is now available. In parallel, the final portion of a British cod stomach data set for the western Barents Sea from 1930-1970s was digitized, processed and quality controlled. The two data sets were then merged. The data set has been made openly available and an article on it has been accepted for publication. Based upon the Norwegian-Russian stomach database the trends in cod diet with increasing age/size have been examined and predator-prey interactions and prey-size selectivity assessed. Shifts in diet with increasing cod size were observed, with fish becoming gradually more important prey. Diet varied significantly inter-annually during 1984-2016, consistent with changes in prey species abundance. Differences were also observed between seasons, with capelin generally dominating the diet in winter, whereas cod, polar cod, and other small fish species were prevalent in the summer/autumn months. This work has been published in ICES Journal of Marine Science and presented at several occasions. In CoDINA we have also studied variability patterns in cannibalism among cod in the Barents Sea. The database is here used to investigate how cannibalism fluctuates with the size distribution of cod. This work is accepted and soon ready for publication. In a fourth manuscript we explore how the expansion of the cod's living area to the north and east affects the ecosystem in the NE Barents Sea and especially to what extent cod has contributed to the observed decline of the arctic fishes there. The first results on this were presented at the 18th Norwegian-Russian fisheries science symposium in Murmansk in 2018 and the manuscript for a scientific article is now being finalized. The amount of snow crabs in the Barents Sea has increased a lot during the 2000s. Analysis of the cod stomach data also show a clear increase in the amount of snow crab consumed by larger cod. This is an example of how information sampled by cod may be applied to make indicators of ecosystem state. This work is accepted for publication in ICES Journal of Marine Science. At CEFAS, UK, work has been done on developing, customizing, and testing so-called Ecopath models to our purposes. A manuscript on this has been written. The University of Sheffield has a research fellow, funded by NERC, who is working closely with CoDINA. He is developing and applying a climate-based distribution model for cod and their prey in the Barents Sea (a so-called maximum entropy model). The stomach database is used to verify the model results.

Post doc Rebecca E. Holt, har fått fast prosjektfinansiert stilling ved IBV, UiO. Hun har fått økt erfaring med både dataanalyse, skriving og internasjonalt samarbeid. Prosjektleder Geir Ottersen har fått videre erfaring med prosjektledelse. De russiske prosjektdeltagerne har fått videreføre samarbeidet med Havforskningsinstituttet. Dette er bra for kunnskapsutveksling på forskernivå og har hjulpet med å opprettholde båndene mellom institusjonene. Joseph Molloy, Universitetet i Sheffield, fikk PhDstipend fra det britiske forskningsrådet NERC for arbeid tett opp mot CoDINA. Bakgrunnen for dette er at hans veiledere er samarbeidspartnere i CoDINA. Forskerkollegaer kan arbeide videre med et stort datasett som har blitt gjort offentlig tilgjengelig og dokumentert. Indirekte kan resultater fra prosjektet bidra til forbedret grunnlag for rådgiving til fiskeri- og miljøforvaltningen. Prosjektet har også bidratt til internasjonalt og tverrfaglig samarbeid.

CoDINA aims to increase the understanding of pelagic food web dynamics in sub-Arctic ecosystems. This we will do by studying the interaction between Barents Sea cod and its prey. We will examine the mechanisms generating predator- prey interactions to gain a deeper understanding of trophic connections in the BS. We are interested in how these may change within the annual cycle, between years and spatially as a consequence of dynamics internal to a single population or between populations and how the functionality is influenced by external pressures like climate and fishing. Cod have been important predators in many ecosystems in the North Atlantic, but their ecological significance has generally diminished with reduced abundance. In contrast, the cod stock in the BS is at a record high level. This is a well-studied and data-rich stock, but still large holes remain in our knowledge. Some of these knowledge gaps are related to how the function of cod in the ecosystem may vary in space and time and how this may be detected in spatiotemporal patterns in their diet. While earlier studies mainly have built on data from the 1980s and onward, we in the proposed project have a unique opportunity to also include the 1930s-1960s, including a warm (until the mid 1950s), but until now data poor period. Newly digitized Cefas diet data for BS cod from this period will be used in unison with data from the joint IMR-PINRO stomach content database (1984-), the PINRO field feeding analysis data (1947-1985), population and hydrographic data from IMR,PINRO ICES and other sources. Based upon these data and the project team's joint expertise in a variety of analysis and modelling approaches we anticipate to move significant steps forward towards successful modelling, understanding and (ecosystem-based) management of the Barents Sea cod stock and its main prey species. We further expect that (parts of) the knowledge provided will be transferable to other marine sub-Arctic systems.