Introduction
Environmental impact from sea-based finfish aquaculture is becoming increasingly prioritized in the authorities handling of the industry. In Norway, this has led to requirements for collection of sludge from Atlantic salmon aquaculture sites in addition to the regulations on the near-field effects on the sea bed. In addition, sludge from aquaculture is increasingly viewed as a resource towards other products such as fertilizers, feed and energy. However, the amount and quality of sludge being produced and available for collection from sea-based aquaculture is not sufficiently investigated. In this work state of the art on mass-balance calculations for feces production for Atlantic salmon, and properties of feces and feed spill is reviewed before drift of feces and feed spill is analysed. Thereby, collectable proportion of dry weight is estimated.
Methodology
A literature review based on a structured search mixing relevant keywords such as sludge, collection, aquaculture, marine, feces, feed spill, sinking/sinking velocity, mass balance, salmon, waste, pellets, current, dispersal, digestibility, and more was performed. 62 scientific papers and reports were found to be relevant for Atlantic salmon fish farming, and were divided into three categories: (i) Mass balance, (ii) Feces and feed properties, and (iii) Collection of feces and feed. Quality of experiments, consensus between researchers and publications, and knowledge gaps were assessed to understand the status of current knowledge.
Based on knowledge from the literature review, analyses on feces and feed spill drift in sea were performed. Understanding how feces and feed spill travel after they are released into the sea in the net pen is crucial to estimate collection rates for sludge collection systems. Two approaches were combined; First a limited number of analyses using Computational Fluid Dynamics (CFD) were performed for variations in drop points, densities, drag coefficients, settling velocities and current speeds. Then, a simplified model was tuned to match the results of the CFD-analyses as closely as possible. A series of scenarios for aquaculture production and environmental conditions were then assessed using the simplified model.
Results and discussion
Mass balance is an established concept for quantifying inputs and outputs in aquaculture systems and is frequently extended to include feed waste. This approach enables the establishment of critical relationships between feed input, fish growth, and production contributions. For Atlantic salmon, mass balance formulas are generally based on the quantity of feed consumed, the amount digested, and the nutrients retained within the fish. Consequently, feed that remains unconsumed is classified as feed spill, undigested material as feces, and non-retained nutrients as excretion.
While the literature provides numerous parameter values for these calculations, they are often inconsistent, case-specific, and difficult to validate for novel applications. This poses a significant challenge for accurate estimation and strategic planning. Digestibility is inherently dependent on feed composition and varies across different nutritional components, with some studies also indicating variations linked to fish size. Furthermore, the effect of environmental factors such as temperature introduce additional complexity to these estimates.
Fecal composition depends on feed ingredients and fish digestion, though it does not appear to exhibit substantial variation. In contrast, fecal shape, size, consistency, and settling velocity appear to vary more significantly depending on fish size, feed types, and the potential inclusion of binders. Fecal content typically remains within defined intervals, with settling velocities falling within these specific ranges.
Simulations of particle dispersion and the capture efficiency of a collection funnel were therefore conducted using various parameter settings, including different sizes, drag coefficients, and weights. Particles were released from multiple areas within the cage to represent the spatial distribution of a fish population. A collection funnel was placed in two different positions beneath the cage, and the percentage captured was measured across varying current velocities.
The CFD analysis demonstrates that results range from 0.0% to 97.4% capture efficiency within reasonable variations in the experimental setup. The simplified model demonstrated a strong capability for adaptation and replication of the CFD results by adjusting the flow velocity reduction through the net wall. However, an inherent inaccuracy remains, as the model produces both higher and lower values than the CFD results for different parameter choices at the same reduction factor. This is likely due to the uniform flow field in the simplified model, whereas the CFD analysis accounts for significant spatial variation—specifically, that current velocity is higher beneath the cage than inside, and that it decreases toward the aft of the cage.
Combining these results allows estimating collectable emissions for concrete setups for sludge collection. This gives valuable insight into the true efficiency of mitigation measures for particulate emissions and environmental impact from fish farms
Acknowledgment
This research is part of the ongoing project "Calculation model for emission and collection of sludge from fish farms at sea", financed by the Norwegian Seafood Research Fund (project number 910382).