The aquaculture industry on the Faroe Islands has been expanding for many years, and new suitable locations are harder to find. In recent years waves and wave-generated currents have become an increasing challenge. A large portion of the fish died in a 2016 storm at the aquaculture site in the bay of Sandur. Similarly, in a 2020 storm, at the aquaculture site at Froðba, all the fish in most of their 14 cages died, with an estimated loss to be in the order of one million salmon. In both of these recent extreme events, the locations of the installations were in the outer part of the fjord close to sheltering shoals, where waves were breaking and generated jetty currents.
The 50-year design wave-height Hm0,50 on the shelf region outside the shoal at Sandur is estimated to be 16m–18 m. Behind the shoal, the most significant recorded Hm0 value during the storm was 4 m, even when offshore measurements were in the order of Hm0,50.
The site avoided the waves’ direct effects; instead, strong currents generated by breaking waves hit the area.
Fiskaaling has measured the waves and currents at several locations in the bay of Sandur during the last five years. Based on these measurements, the farming cages are moved farther in the bay.
A numerical model
The challenge is to locate the path of the jet current to ensure not to place the cages in it. You can measure wave height and current velocity at a specific site in all weather, and you can measure wave height and current velocity all over the area in calm weather. But, you must have a numeric model to predict wave height and current velocity in the hole area in all weather. You do also need a numeric model to predict the 50-year design storm.
In the PhD project at hand, we will apply suitable models and use the data from Sandur to determine the best model configuration for consultancy tasks of this type. Compared to similar studies of waves and wave-generated currents, the Sandur case is somewhat unique due to the wave severity, deeper settings and the link to high-resolution fish-net movements. Data include wave-buoy measurements on the shelf, wave-buoy measures, including ADCP current measurements, in the bay, bottom-mounted ADCPs at several locations in the bay, some with wave measurements, wave measurements with pressure sensors in a line inwards the shore, and measures of movement of the cage based on pressure sensors on the net. The net movement measurements sampled at 1 Hz show time variations corresponding to both the waves’ phase- and group velocity. We will also present initial results from model simulations if all goes according to plan.