Decreasing body size has been proposed as a universal response to increasing temperatures. The physiology behind the response is well established for ectotherms inhabiting aquatic environments: as higher temperatures decrease the aerobic capacity, individuals with smaller body sizes have a reduced risk of oxygen deprivation. However, empirical evidence of this response at the scale of communities and ecosystems is lacking for marine fish species. Here, we show that over a 40-year period six of eight commercial fish species in the North Sea examined underwent concomitant reductions in asymptotic body size with the synchronous component of the total variability coinciding with a 1-2 °C increase in water temperature. Smaller body sizes decreased the yield-per-recruit of these stocks by an average of 23%. Although it is not possible to ascribe these phenotypic changes unequivocally to temperature, four aspects support this interpretation: (i) the synchronous trend was detected across species varying in their life history and life style; (ii) the decrease coincided with the period of increasing temperature; (iii) the direction of the phenotypic change is consistent with physiological knowledge; and (iv) no cross-species synchrony was detected in other species-specific factors potentially impacting growth. Our findings support a recent model-derived prediction that fish size will shrink in response to climate-induced changes in temperature and oxygen. The smaller body sizes being projected for the future are already detectable in the North Sea.

No abstract

The assumption that spawner biomass is directly proportional to total egg production by fish stocks underlies most spawner-recruit relationships. Despite its importance, this assumption is largely untested. Data describing the reproductive potential of Northeast Arctic (NEA) Atlantic cod (Gadus morhua) were used to test the proportionality assumption for a period showing strong variation in condition and abundance of spawners. At the individual scale, relative fecundity varied with length and condition. This decreases the likelihood that the proportionality assumption is valid. At the stock level, total egg production was estimated (1985-1996) using acoustic estimates of total abundance and demographic information from trawl sampling in combination with year-specific fecundity-length relationships. For NEA cod, spawner biomass estimated by virtual population analysis (VPA) was not proportional to total egg production. Compared with VPA-based estimates of spawner biomass, total egg production is an improved index of recruitment potential because (i) the magnitude of variation observed in total egg production was closer to that observed in recruitment and (ii) the relationship between recruitment at age 1 and total egg production approaches the origin. The results suggest that further progress in explaining recruitment variation will be achieved using more sensitive measures of the true reproductive potential of the stock.

No abstract

Ocean climate impacts on survivorship and growth of Atlantic salmon are complex, but still poorly understood. Stock abundances have declined over the past three decades and 1992-2006 has seen widespread sea surface temperature (SST) warming of the NE Atlantic, including the foraging areas exploited by salmon of southern European origin. Salmon cease feeding on return migration, and here we express the final growth condition of year-classes of one-sea winter adults at, or just before, freshwater re-entry as the predicted weight at standard length. Two independent 14-year time series for a single river stock and for mixed, multiple stocks revealed almost identical temporal patterns in growth condition variation, and an overall trend decrease of 11-14% over the past decade. Growth condition has fallen as SST anomaly has risen, and for each yearclass the midwinter (January) SST anomalies they experienced at sea correlated negatively with their final condition on migratory return during the subsequent summer months. Stored lipids are crucial for survival and for the prespawning provisioning of eggs in freshwater, and we show that under-weight individuals have disproportionately low reserves. The poorest condition fish ($ 30% under-weight) returned with lipid stores reduced by $ 80%. This study concurs with previous analyses of other North Atlantic top consumers (e.g. somatic condition of tuna, reproductive failure of seabirds) showing evidence of major, recent climate-driven changes in the eastern North Atlantic pelagic ecosystem, and the likely importance of bottom-up control processes. Because salmon abundances presently remain at historical lows, fecundity of recent year-classes will have been increasingly compromised. Measures of year-class growth condition should therefore be incorporated in the analysis and setting of numerical spawning escapements for threatened stocks, and conservation limits should be revised upwards conservatively during periods of excessive ocean climate warming.