SEAwise Report on improved predictive models of growth, production and stock quality under different habitat scenarios and incorporating experimental results
The SEAwise project works to deliver a fully operational tool that will allow fishers, managers, and policy makers to easily apply Ecosystem Based Fisheries Management (EBFM) in their fisheries. Understanding how ecological drivers impact stock productivity through growth, condition and maturity is essential to advance towards ecosystem-based fisheries management. This SEAwise report presents final predictive models of fish growth, condition and maturity in each of our four regional case studies. The biological processes fish growth, condition and maturity were studied in terms of body size, condition factor, otolith increments and size at first maturity, and, in laboratory studies, in terms of physiological, metabolic and microbiota parameters. The data ranged from individual fish, to sampling, haul or stock level. Accordingly, the methods varied according to the specific features of the process under study and the available data.
In the Baltic Sea, two regimes were identified in the weight-at-age time series of herring in the Gulf of Riga (1961-1988 and 1989-2020). During the first period the main driver of the individual annual growth of the fish was the abundance of the copepod L. macrurus macrurus, while the pattern was less clear during the second period. Summer temperature was not a significant driver of the individual growth. Mediated length-based growth models and linear mixed models applied to weight-at-age data of six stocks showed forecast potential for only one stock (sol.27.20-24). The most plausible model estimated positive impacts of temperature, salinity and the interaction between weight and salinity.
In the Mediterranean Sea, the first group of analyses focused on effects on size at first maturity, condition factor and growth in South Adriatic Sea (GSA 18) and North-West Ionian Sea (GSA 19). In most cases, the most significant environmental driver was bottom temperature, although some relationships with bottom salinity and primary production were also found. Analyses of otolith increments of red mullet were used to investigate differences between areas. The best forecast model included effects of surface temperature, bottom temperature, bottom oxygen, longitude and latitude. Forecasted growth under the RCP 8.5 scenario suggested an increase in red mullet growth in the medium term and a decrease in the long term. Secondly, the analysis of the impact of the environmental variables on hake and red mullet in the Eastern Ionian Sea (GSA 20) pointed to temperature as one of the main drivers. For condition factor, the main factors included temperature, zooplankton and macrobenthos depending on the species, the age groups and the season.
In the North Sea, the performance of a variety of models were assessed in terms of model fit and predictive capability for seven stocks. Cod, saithe, haddock and whiting showed forecast potential for at least one of the methods. For North Sea haddock, all methods with forecast skill included a negative temperature effect, whereas for North Sea saithe and whiting the model without environmental covariates had the most promising forecast potential. For North Sea cod, the most influential variable changed depending on the method employed. Mizer model predictions (package for size-spectrum ecological modelling) with environmental forcing was used to study whether warming in the North Sea is responsible for the low productivity of the cod stock. Time-varying von Bertalanffy model parameters suggested three environmental regimes for the North Sea. Finally, detailed otolith increment analysis was used in the development of growth of sole in the North Sea and in the Irish Sea. In the North Sea, the best model of sole growth included sea bottom temperature, fishing mortality at age, and stock biomass at maturity stage, and their interactions with age and maturity, while in the Irish Sea, the best model included sea bottom temperature and fishing mortality at maturity stage and its interaction with maturity. These results showed a positive effect of temperature on adult growth. However, temperature had unexpected negative effects on juvenile growth, which requires further study.
In the Western Waters, the first study analysed length at age of four species based on Irish groundfish survey data. Across areas, temperature had a negative effect on all age groups for whiting, haddock, and megrim, with more variable impacts observed in blue whiting. Other factors like the Atlantic Multidecadal Oscillation (AMO), fishing mortality, biomass and recruitment also affected length-at-age, although with the direction depending on the species. In the second study, a variety of models were applied to sixteen stocks in the Western Waters (five in the Irish Sea, six in the Celtic Sea and five in the Bay of Biscay). In the Irish Sea, four out of five stocks showed forecast potential. No predictors were included in the best method for the Irish Sea haddock stock, while the best method selected SSB, temperature and some interactions for cod, salinity, temperature and some interactions for sole and SSB, salinity and some interaction for whiting in the area. In the Celtic Sea, three out of six stocks showed forecast potential. However, the selected growth drivers differed between stocks and methods. In the Bay of Biscay, two out of the five stocks studied showed forecast potential. The best method for sardine in Biscay included the predictors SSB, salinity and temperature and some interactions, while the best method for Iberian sardine did not include any predictor. The third study in the Western Waters analysed relative condition factor in the Bay of Biscay and the Celtic Sea and found significantly decrease over time for 12 species, while two species showed an increase, and three species showed no significant trends. The impact of density dependence on body condition was significant for all species except two. Chlorophyll-a and SST was correlated with body condition of some species. However, the direction of the correlations found was different across species. Length and weight of anchovy in the Bay of Biscay declined in the fourth study. For juveniles the best model featured density-dependent effects, while for adult age classes the best model included temperature, resource availability and density-dependent effects. Finally, a temperature dependent growth model for European seabass showed higher growth increments at higher temperature, implying a faster growth for seabass in the Bay of Biscay, compared to the English Channel, Celtic Sea and North Sea.
Beyond the regional case studies, a range of experimental studies were conducted to analyse the impact on European seabass of changing environments. The impact of ocean warming and acidification on growth, endocrine, intestinal, and metabolic parameters, of ocean deoxygenation on metabolic rates, of ocean acidification on sexual maturation and of contaminants and pollutants on bioaccumulation, growth and physiological functions were studied. Data analyses are still in progress, but some of the preliminary results suggested higher thermal growth and extra energetic needs under warming and acidification, no impact of hypoxia on growth and earlier maturation in high acidification scenario. Regarding contaminants and pollutants, PFAS contamination did not affect growth but results suggested higher metabolic costs under chemical and/or climatic stress conditions.
The predictive ability is one of the most important features of a model and was evaluated for some models using a hindcasting approach. The accuracy of the model predictions was measured in terms of the mean absolute error and compared to the accuracy of a naïve approach based on the last three years average. In some of the cases, the model selection was included within the hindcasting approach, whereas in the other cases the prediction skills were assessed on the already selected models. Overall, the results indicated that the best models in terms of model fit did not always have good forecast skills. The improved predictive models developed in this task are readily available to be transferred to WP6.
Read more about the SEAwise project at www.seawiseproject.org