Predicting oxygen levels in Atlantic salmon (Salmo salar) sea cages

Abstract

Dissolved oxygen (DO) concentration is an important factor affecting fish welfare and growth in Atlantic salmon sea cages. As an operational prediction tool, and to aid our understanding of how DO concentrations are affected by cage size, shape and design interacting with the fish biomass and environmental conditions, we have developed a mathematical model based on the advection-diffusion equation for 3D estimates of DO levels. The model requires input of farm geometry, ambient oxygen levels, current speed and direction, feeding rates, fish distribution and biomass statistics.

The model has been tested for Salmar's Ocean Farm 1, a large production unit dimensioned for 1.5 million fish and designed for coping with exposed environmental conditions, and performs well in comparison with DO measurements, particularly at low to moderate current speeds. Although the model produces realistic outputs with the simple inputs that are available, detailed information about fish behaviour and current conditions within the cage is likely to improve model accuracy.

In addition to being a useful tool for better monitoring of oxygen conditions in fish farms, the present model can, given the appropriate inputs, be used as a forecasting tool for predicting the risk of hypoxic conditions in cages, and to evaluate the risks of hypoxic conditions in new types of open, semi-closed or closed production systems.