Abstract

Climate change is commonly associated with many species redistributions and the influence of other factors may be marginalized, especially in the rapidly warming Arctic.The Barents Sea, a high latitude large marine ecosystem in the Northeast Atlantic has experienced above‐average temperatures since the mid‐2000s with divergent bottom temperature trends at subregional scales.Concurrently, the Barents Sea stock of Atlantic cod Gadus morhua, one of the most important commercial fish stocks in the world, increased following a large reduction in fishing pressure and expanded north of 80°N.We examined the influence of food availability and temperature on cod expansion using a comprehensive data set on cod stomach fullness stratified by subregions characterized by divergent temperature trends. We then tested whether food availability, as indexed by cod stomach fullness, played a role in cod expansion in subregions that were warming, cooling, or showed no trend.The greatest increase in cod occupancy occurred in three northern subregions with contrasting temperature trends. Cod apparently benefited from initial high food availability in these regions that previously had few large‐bodied fish predators.The stomach fullness in the northern subregions declined rapidly after a few years of high cod abundance, suggesting that the arrival of cod caused a top‐down effect on the prey base. Prolonged cod residency in the northern Barents Sea is, therefore, not a certainty.

Highlights

  • We examined the influence of food availability and temperature on cod expansion using a comprehensive data set on cod stomach fullness stratified by subregions characterized by divergent temperature trends

  • Generalized linear mixed models were used to analyze how cod occupancy (CO), measured as the proportion of trawl sets with cod present, and cod stomach fullness (CSF) in each of the eleven subregions changed over time

  • The recent expansion of cod in the Barents Sea represents the northernmost occurrence of this species

Read more

Summary

| Sampling procedure and data

The survey was initiated in 2004 and has been conducted annually in August–September (duration 39–57 days)—the time of the year when, on average, the Barents Sea has minimal ice cover. | 3 temperature measurements were obtained from the deployment of a CTD (conductivity, temperature, and depth) probe adjacent to each trawl set in the grid. The subregions were defined by interpolating the spatial variation for large cod and temperature on a joint grid for each year using an additive model with a bivariate tensor product smooth of the spatial coordinates, and a smooth term for depth (Wood, 2006). We binned the Barents Sea into subregions using 4 × 4 and 5 × 5 geographical grid designs in order to assess the robustness of our approach (Table S5)

| Study design
Findings
| DISCUSSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call