Flow dynamics and turbulence in large-scale fish farm net pens

PhD-student: Kristbjörg Edda Jónsdóttir


Associate Professor Jo Arve Alfredsen (NTNU)
Associate Professor Morten Alver (NTNU/SINTEF Ocean)
Senior Research Scientist  Zsolt Volent (SINTEF Ocean)

Duration: Q3 2016 – Q3 2020

The aim of this project is to conduct new theoretical and empirical research on the dynamics of water flow and turbulence inside large-scale net pens.

In order to obtain a new understanding of sea currents that infiltrate large-scale net pens, measurement campaigns have been carried out at two localities in the Lofoten islands, and one offshore Hosnaøyan in Trøndelag. Many sensors have been deployed as part of this campaign, and hopefully they will enable us to shed some light on how sea currents flow through the net pens. This work is still ongoing.

In order to investigate flow on the inside of the pens, it has also been necessary to study the sea currents in the waters surrounding them. In this connection, it was of interest to look into how it is possible to evaluate sea currents in a fish welfare context.

Thus, data from the three “EXPOSED” buoys, as well as two buoys deployed in connection with the RACE project, have been evaluated, taking fish swimming capacity into consideration. In this case, studies focused specifically on the critical swimming velocity of small post-smolt fish and lumpfish.

Results indicate that even if the salmon can adapt to most localities, the lumpfish struggle because the sea current speeds commonly exceed their critical swimming velocities.

Results to date:

  • We have almost finalised the article addressing the assessment of current strength at the localities in a fish welfare context.
  • The experiments at Fornes, Josommarset and Hosnaøyan have been completed.

Status and further work:

  • Further measurement campaigns are planned at Hosnaøyan and Fornes.
  • Testing of measurement methods will be carried out at Hosnaøyan in the spring of 2019.
  • A continuation of the first article will be carried out using a proposed method on model data.