Project design and implementation
The work is to be done in two work packages (WPs).
WP1: Large scale demonstration and trial of “Well” cages
Commercial skirt vs “Well” cages
At the commercial salmon farm Haverøy with no naturally occurring brackish water layer, 4 designated trial cages will be used (160 m circumference): Two cages set up as traditional skirt cages and two cages with “Well” setup, and monitored for most of a production cycle from November 2017 to October 2018 (about 1.5 kg to harvest).
All cages will have 6 m deep outer tarpaulin skirts installed and hold lice-eating cleanerfish and salmon in equal numbers. “Well” cages will contain deep lights along with an impermeable central inner tarpaulin of 6 m depth and 15 m diameter containing the freshwater lens. Inside the freshwater lens a dimmable 400 W light will be placed at 1 m depth, and can be used to attract fish into the fresh-to brackish water layer. Specifically, “Well” cages are lit either inside the freshwater lens or below 10 m depth, which in both cases will contribute to reduce the infection pressure by maintaining salmon either in fresh- to brackish water or in deeper water and less lice abundant water, respectively.
The inner tarpaulin will be continuously filled with freshwater from a reverse osmosis desalination unit (RO) positioned at the feed barge, which will create a fresh- to brackish water gradient (salinity of 5–10 in upper part, with maximum salinity of 20 except in the prevalent mixed zone in the lower end of the inner tarpaulin). Water intake depth is made flexible, thus the FW-lens will always contain water of the temperature most optimal to Atlantic salmon (closest to 15 °C) in order to stimulate FW-lens usage.
Desalinated water at 0,5–3 ppt will be pumped at a rate of 600–800 m3/day/pen leading to a water exchange rate of some 50 percent per day in the central inner tarpaulin. Depending on oxygen conditions the water exchange rate may be increased or oxygen will be provided through super-oxygenation of the added RO-water.
Continuous real time environmental logging systems will be implemented in all 4 trial cages. Each “Well” cage will be monitored with respect to temperature, oxygen and salinity at 1, 3 and 5 m depth inside the FW-lens. Additionally, the same variables will be monitored outside the FW-lens at 1 m depth, to compare temperature and oxygen content within and outside the FW-lens, and to monitor salinity outside the FW-lens to see whether or not this will also increase by freshwater escaping during rough weather. Both control cages will have continuous monitoring of temperature, oxygen and salinity at identical placement as outside the FW-lens in “Well” cages. For additional data on salinities experienced by fish, salinity histories of 12 “sentinel fish” in the two “Well” cages will be tracked by fitting them with acoustic salinity tags and receiving signals at an upper and lower receiver in each cage. This will be repeated in every season so salinity data is gathered throughout the commercial trial.
The tarpaulin- and mooring design will be optimized during the production cycle if needed, but will at start-up have a cylindrical shape. It will be weighed down with a specially made bottom ring to maintain stability of the FW-lens in rough weather, and also consist of completely impermeable material to prevent dilution by osmosis. Contingency tarpaulin designs have been made, and if the cylindrical shape cannot create a volume of stable fresh- to brackish water, we will produce and test the contingency designs. Equipment for current measurements are placed in close proximity to “Well” cages, and will measure profiles of current speed in real time in the 0 and 40 m depth range. These data will be correlated with real time salinity monitoring within the FW-lens to establish a relationship between current speed (m/s) and achievable salinity with max freshwater production.
Feed will be delivered via a standard surface feed spreader in all cages. However, the air pressure used in the surface spreaders within “Well” cages will be adjusted so the feed is not spread beyond the inner skirt diameter, thereby increasing attraction to freshwater. Feeding intensities can be varied and thereby we can adjust the depth at which the pellets reach salmon, thus making the inner tarpaulin act as a de-facto deep feeder: Fish must descend to 6 m to forage if not in the FW-lens itself. This ensures that feeding takes place either in fresh- to brackish water or at depths below those where infective stages of salmon louse are most abundant, both of which will have a positive effect on infection rate. Feeding in the upper 6 m area outside the FW-lens will not occur.
Lice levels and fish welfare will be recorded every 3–4 weeks depending on lice development times at different temperatures (scoring sheet from project “Kunnskapssammenstilling om fiskevelferd for laks og regnbueørret i oppdrett (FISHWELL)”, FHF-901157
). Environmental conditions at a reference location will also be continuously monitored throughout the year-long trial to explain variations in lice levels, fish welfare and fish distribution. Sampling and behavioural observations of cleaner fish will be carried out to understand how their welfare is affected and what their lice consumption rates are.
WP2: Testing brackish water effects on attached salmon lice
Brackish water survival of attached copepodids
At the Institute of Marine Research (IMR) Matre Research Station, 16 tanks with automated temperature and salinity regulation will be used over 2 periods of 1 month. Tanks maintained at a salinity of 34 and temperature of 14 °C with salmon in each will be subjected to a dose of 30 salmon lice copepodids per fish. Once reaching one day post infection (PI), groups of 30 fish with attached copepodids (comprising 10 fish from each of the three holding tanks) will be transferred to either a seawater control tank or a low salinity treatment tank for 1, 3, 6, 12, 24 or 48 hours or combinations of repetitive exposures based on observations from sentinel fish in WP1 (e.g. 10 minutes every 2 hours; 1 minute per hour for 1–7 days). After treatments, the fish will be transferred again to seawater recovery tanks until 8 days PI, when lice are expected to develop into chalimus 1 and 2, as performed in a previous study. At this point, the fish will be lethally sampled and counted for lice to assess low salinity exposure effects compared to the control. These procedures will be repeated for minimum 4 brackish water salinities between 0 and 20.