A Dynamic Energy Budget (DEB) model for simulation of growth and bioenergetics of blue mussels (Mytilus edulis) has been tested in three low seston sites in southern Norway. The observations comprise four datasets from laboratory experiments (physiological and biometrical mussel data) and three datasets from in situ growth experiments (biometrical mussel data). Additional in situ data from commercial farms in southern Norway were used for estimation of biometrical relationships in the mussels. Three DEB parameters (shape coefficient, half saturation coefficient, and somatic maintenance rate coefficient) were estimated from experimental data, and the estimated parameters were complemented with parameter values from literature to establish a basic parameter set. Model simulations based on the basic parameter set and site specific environmental forcing matched fairly well with observations, but the model was not successful in simulating growth at the extreme low seston regimes in the laboratory experiments in which the long period of negative growth caused negative reproductive mass. Sensitivity analysis indicated that the model was moderately sensitive to changes in the parameter and initial conditions. The results show the robust properties of the DEB model as it manages to simulate mussel growth in several independent datasets from a common basic parameter set. However, the results also demonstrate limitations of Chl a as a food proxy for blue mussels and limitations of the DEB model to simulate long term starvation. Future work should aim at establishing better food proxies and improving the model formulations of the processes involved in food ingestion and assimilation. The current DEB model should also be elaborated to allow shrinking in the structural tissue in order to produce more realistic growth simulations during long periods of starvation.

Highlights ► Particle retention efficiency (RE) of natural seston was investigated in M. edulis. ► The smallest particles shown to be effectively retained (100%) were close to 7 μm. ► RE most often increased progressively from small to large particles. ► Maximum RE was reported at 30 to 35 μm in 8 of 12 experiments. ► Temporal changes in particle size distribution coincided with changes in RE maxima.

No abstract

my supervisors; Øivind Stand, Peter Cranford and Svein Rune Erga that have spent much time in directing and improving the thesis. Øivind led the project and his advices, clear mind, overview and always positive attitude made the fulfillment of this PhD not only possible but to an enjoyable task. Peter has guided me through the bivalve eco-physiology and the experimental methodology, providing vital advices and comments. Svein Rune has supported me from day one in this task. Thanks! I would like to thank my co-authors for their contribution in the papers; Øivind Strand, Jan Aure, Peter Cranford, Arne Duinker, Tonje Castberg, Asbjørn Svardal and Cathinka Krogness. The extent of the eco-physiological measurements would not have been possible without the commitment of Cathinka Krogness. This thesis was improved as a result of suggestions, comments and discussion with Jan Aure, Jon Grant, Aad Smaal, Cedric Bacher, Marianne Alunno-Bruscia, Stein Mortensen, Raymond Bannister and Rune Rosland. The CANO project in general and specifically the CANO bivalve team provided a dynamic environment in which small and larger challenges have been addressed. Thanks to all the 5

Blue mussel Mytilus edulis cultures contribute to nutrient cycling in coastal ecosystems. Mussel populations filter particulate nutrients from the water column and inorganic nutrients are regenerated by excretion of metabolic wastes and decomposition of (pseudo-)faeces. The objective of this study was to determine the intra-annual variability in nutrient regeneration by mussel rope cultures in oligotrophic fjord systems. In situ respiration and nutrient uptake and release rates of 1 m mussel ropes were measured using 250 l pelagic chambers. There was a 20-fold difference between winter and summer respiration and nutrient release rates. Inorganic nitrogen release ranged from 50 to 1000 µmol h -1 per meter rope. These variations were mainly related to mussel growth but were also related to changes in water temperature and biofouling biomass (organisms that colonized the mussel ropes). In total 24 genera of fouling organisms were observed, diversity increased over time (ranging from 2 to 12 genera m -1 ), and fouling biomass was mainly characterised by ascidians (max. 37 ± 14 g m -1). However, mussels dominated the culture ropes, representing > 90% of total faunal biomass. The amount of organic material associated with the ropes was stable (6.9 ± 0.3 g m -1). At the scale of one mussel farm, nutrient regeneration by mussel rope cultures increased inorganic nitrogen concentrations by 20% and inorganic phosphorus concentrations by 5% during summer conditions. During winter, there was no significant effect of mussel cultures on the inorganic nutrient pools. Nutrient regeneration by mussel cultures also affected stoichiometry as nutrients were excreted in dissimilar proportions (nitrogen > phosphate > silicate). The increased nutrient availability may contribute to primary production, especially in nutrient-limited (oligotrophic) fjord ecosystems. However, fjordscale effects are largely dependent on hydrographic conditions of the fjord system.