Technical University of Denmark
SEAwise_D3_4_growth_w_doi.pdf (1.02 MB)

SEAwise Report on improved predictive models of growth, production and stock quality.

Download (1.02 MB)
online resource
posted on 2024-01-24, 11:22 authored by Leire Ibaibarriaga, N. Aldanondo, S. Alfonso, P. Álvarez, Luke Batts, Isabella Bitetto, G. Boyra, Mollie Elizabeth BrooksMollie Elizabeth Brooks, T. A. Bui, Pierluigi Carbonara, Guillem Chust, U. Cotano, M. De Troch, J. (Jochen) Depestele, Paul Dolder, M. Donnaloia, Heli Einberg, M. Erauskin-Extramiana, J. A. Fernandes, A. Fontán, Paul Gatti, Kristiina Hommik, Xabier Irigoien, Bernhard Kühn, Christophe Lebigre, C. Loredana, D. Mazurais, Paco Melia, C. Neglia, Stefan NeuenfeldtStefan Neuenfeldt, Henn Ojaveer, M. Palmisano, Dimitris Politikos, J.J. (Jan Jaap) Poos, I. Putnis, G. Rubene, M. Santos, Marie Savina-Rolland, A. Servili, L. Simonet, M. Spence, F. G. Taboada, Marc Taylor, Robert Thorpe, Konstantinos Tsagarakis, Morgane Travers-Trolet, Andrés Uriarte, Celia Vassilopoulou, Woillez, Mathieu, Walter Zupa, Anna RindorfAnna Rindorf

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 and understanding how ecological drivers impact stock productivity through growth, condition and maturity is essential to this proces. In this SEAwise report, we present the predictive models of fish growth, condition and maturity obtained so far in each of the four regional case studies.

The biological processes (fish growth, condition and maturity) were studied in terms of body size (weight-at-age, length-at-age), condition factor, otolith increments and size at first maturity. Underlying data were available at different levels, ranging from individual fish, to sampling haul or stock level. Accordingly, the methods employed varied across case studies to adapt to the specific features of the process under study and the available data.

The methodology encompassed statistical models (linear models, generalised additive models, mixed models, Bayesian nested hierarchical models, changepoint models), otolith growth increment analyses and mechanistic models (DEB-IBM model coupled to the environment and mizer model). Some of these models were focused on detecting overall trends, including potential changepoints along the time series or identification of the main intrinsic factors. Other models explored the impact of ecological drivers such as temperature, salinity, food availability or density dependence.

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 abundance of the adult stages of E. affinis affinis was the dominating explanatory variable affecting herring growth during the second period. Neither SSB nor summer temperature during the main feeding period were significant drivers of the individual growth in the two distinct ecosystem regimes.

In the Mediterranean Sea, the analysis of the impact of the environmental variables on biological parameters like size at first maturity, condition factor and growth in South Adriatic Sea and North-West Ionian Sea showed some significant effects in relation to the different species/area. In most of the cases, the environmental driver was bottom temperature, although some relationships with bottom salinity and primary production were also found. The model outcomes suggested that temperatures prevailing in deeper waters were the most significant factor affecting gonad maturity of hakes, while those in the shallow zone had the main impact on the L50 of red mullets. Condition factor of hake and red mullet in the Eastern Ionian Sea were affected not only by temperature, but also by zooplankton abundance.

In the North Sea, mediated length-based growth models, linear mixed models and state-space linear mixed models were applied to four gadoids, two flatfishes and one pelagic stock and their performances were assessed in terms of model fit and predictive capability. For the mediated length-based growth model approach, the best model differed across stocks, but density dependent mediation effects were significant for five out of the seven stocks. Regarding the linear mixed models, the two types of models and the different penalisation procedures led to different models across stocks. Among the additional ecological variables, surface temperature was the most frequently included in the final model, closely followed closely by SSB and to a lesser extent by NAO. Detailed otolith increment analysis was used in the development of multidecadal biochronologies of average annual growth of sole in the North Sea and in the Irish Sea. In the North Sea, the best extrinsic 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 stage, while in the Irish Sea, the best extrinsic model included sea bottom temperature and fishing mortality at maturity stage and its interaction with maturity stage. These results confirmed the expected positive effect of temperature on adult growth. However, in the North Sea, temperature showed unexpected negative effect on juvenile growth, which might be linked to changes in food availability and/or intraspecific competition and need to be further studied. The mizer model (package for size-spectrum ecological modelling) with environmental forcing was used to study whether warming in the North Sea is responsible for the failure of the cod stock. The simulated fish community response when recruitment and carrying capacity depended on surface temperature fitted better with the assessment data than when the environment was fixed. However, the qualitative differences remain, suggesting that temperature effects were not the main cause of the model-assessment disparity.

In the Western Waters, the mediated length-based growth models developed for the North Sea case study were applied to 14 stocks in the Celtic Sea. The best model differed across stocks, but again SSB mediation was significant for most of the stocks. From visual inspection of the plots, however, it was noted that the raw data from certain stock objects showed a reduced growth compared to the model fits, requiring further analyses. The analysis on biological measurements of individuals collected at fish markets, observers at sea or during scientific cruises allowed to study temporal variations in body size and condition factor of benthic, pelagic and demersal species in the Celtic Sea and the Bay of Biscay. The linear models indicated a significant negative monotonic relationship of sizes at all ages for anchovy and pilchard, but variations in size at age were less clear and significant for benthic and demersal species. In contrast, the results of the body condition indices showed a moderate but significant decrease for all the studied 19 species over time. The in-depth analysis for anchovy in the Bay of Biscay based on research surveys confirmed the decline in the length and weight of anchovy in the Bay of Biscay and pointed to a decline in body condition toward slender body shapes. Detected associations between temperature and size became more apparent for adult age classes than for juveniles, whereas the association between anchovy size and the biomass of spawners was more important for juvenile than for adult age classes. Associations between anchovy size and chlorophyll-a concentration were in general weak. Finally, the DEB-IBM model coupled to the environment that is under development for the two main seabass stocks of the North East Atlantic will provide further insights on how growth, condition and maturation can affect the future dynamics and productivity of these stocks.

Read more about the project at


Shaping ecosystem based fisheries management

European Commission

Find out more...


ORCID for corresponding depositor

Usage metrics

    DTU Aqua



    Ref. manager