Hans Tobias Slette at SINTEF says that now that the fish farming sector is looking to operate with bigger facilities and more fish in each net pen, reliable calculations of oxygen availability are more important than ever. Here we see him demonstrating a so-called miniDOT, which is a device for measuring oxygen. Photo: Daniel Gløsen

Researchers want to create a forecasting system for oxygen concentrations in sea water

Farmed fish suffer if there is too little oxygen in the water. A system that can display oxygen concentrations may make it easier to supply this essential gas if the water becomes oxygen-poor.

All fish are dependent on the dissolved oxygen in sea water. To little oxygen in a farm net pen will result in loss of appetite and reduced fish welfare. A net pen is the ‘cage’ suspended in the sea in which the fish are kept.

It thus becomes a problem if a farm operator finds out that oxygen concentrations are consistently too low. So, why is this happening?

“One factor, of course, is that the fish are constantly consuming the oxygen”, says Ingrid Ellingsen, who is a research scientist at SINTEF. “This causes concentrations of the gas to be reduced, especially if the net pens are experiencing poor water throughflow, or if the fish are too densely crowded. But there are also a number of other factors at play”, she says.

Together with her colleague Pascal Klebert and others, she is looking to evaluate how oxygen concentrations are impacted by other factors in the water in and around a fish farm.

Variable supply

The concentration of oxygen in any given area is affected by factors such as the photosynthesis of local plants, which generates oxygen, and the breakdown of biomass, which consumes the gas. 

Respiration by other fish also plays a role, as do other factors such as sea temperature, currents, wave action, tidal cycles, the supply of freshwater by local rivers and oxygen exchange between the sea and the atmosphere. All of these factors ultimately influence the supply of oxygen to the surface waters in the net pens.

In other words, the researchers have a complex set of factors to get their heads around.

SINTEF holds a number of aquaculture research licences along the coast of Trøndelag. These enable research teams to carry out experiments under real-life conditions in what they call ‘full-scale marine laboratories’.

The laboratories practice routine farming operations in parallel with taking measurements and running research experiments. In the long term, the researchers hope to make it possible to develop new technologies that will resolve some of the many problems facing the aquaculture industry.

As part of the research projects called ROXY and Oxygen Buholmen, researchers from SINTEF Ocean are looking into oxygen availability for farmed fish.

Senior Research Scientist Ingrid Ellingsen works at SINTEF Ocean’s Fisheries and New Biomarine Industries department. Photo: Daniel Gløsen

A very relevant issue

“We know that low oxygen concentrations are a very relevant issue for many operators”, says Ellingsen. “Not least during periods involving low water throughput or major algal blooms. Our aim is to provide the sector with the tools it needs to obtain a better understanding of what is happening. Then it will be able to identify the actions it has to take”, she says.

“Now that the fish farming sector is looking to operate with bigger facilities and more fish in each net pen, reliable calculations of oxygen availability are more important than ever”, says Hans Tobias Slette. Slette is an aquaculture researcher and one of Ellingsen’s co-workers.

His job is to carry out detailed investigations in individual pens in the research farms.

“We’re currently using sensors to measure oxygen concentrations at different points in the pens, and using mathematical models to extrapolate data between the sensor locations”, he says. “At the same time as developing this tool, we’re also making control measurements to see how accurate our extrapolation estimates are. Our aim is to find out the minimum number of sensor points we need to obtain sufficient data about the variation in oxygen content within a given net pen”, he says.

This is necessary because a single net pen may exhibit significant variation, and such detailed knowledge is essential if operators are to implement targeted actions.

Reorganisation can address the problems

“If oxygen concentrations are low because there is in fact too little oxygen across a large area of sea on a given day or during a given season, there will be little point in installing pumps simply to replace the water in a single pen”, says Ellingsen.

She believes that such considerations demonstrate the value of establishing a good knowledge base.

For this reason, researchers are seeking out the various causes behind what the sensors are telling them. What factors are acting together during periods when oxygen concentrations are very low?

Nature determines the flow of water through a net pen. But so too does the location of an individual pen within the farm facility. When many collars are packed closely together, water flowing through a given pen may have already passed through other adjacent pens. If the water is already oxygen-poor, the sensors will continue to record oxygen-depleted water.

“Sometimes, based on what we know about oxygen availability, it may be a good idea to reorganise the placement of net pens in a facility”, says Ellingsen. “Operators have a number of options here” she says.

Saving money and reducing suffering

Researchers are currently using a number of different methods in their attempts to study oxygen conditions in the context of the aquaculture sector.

They compare different types of oxygen concentration profiles in the water column with maps of bottom conditions, and combine these with measurements of turbulence, wind conditions, wave and current action, water temperature and tidal cycles.

This enables them to obtain knowledge that can be transferred to operational settings outside the test facilities.

“If we become able to forecast oxygen concentrations in an area before a fish farm is installed, we will save the industry from potentially misdirected investments and the fish from unnecessary suffering”, says Slette.

Ingrid Ellingsen envisages that the database which her team is now accumulating can be used as the basis for developing a forecasting approach comparable with today’s weather forecasting systems.

This article was first published on the website Forskning.no on 16 November 2023.