Thomas A. McClimans

Laboratory simulation of the ocean currents in the Barents sea

Abstract A rotating laboratory model of the Barents Sea was forced by computed inflows of Atlantic Water and Arctic Surface Water for the period 1979–1984. Ad hoc tidal excursions over the shoals north of Bear Island and deep water production as a result of winter cooling and salt rejection in the eastern part of the basin were calibrated in the model. The high spatial resolution in the basin, which was 5 m in diameter, provided the basis for simulating several physical scales simultaneously. The simulated current features of interest include (1) the spreading of the Norwegian Coastal Current over Tromsoflaket, (2) a warm-core jet along the southeastern slope of the Svalbardbanken, which pushes the ice front far to the NE of Hopen Island, (3) the anticyclonic circulation around Sentralbanken, which drives Arctic Surface Water and ice far south in the eastern basin, (4) Norwegian Coastal Water flowing north across the Bear Island Channel, (5) deep water outflows north through the Franz-Victoria Trough and west through the Bear Island Channel, (6) the dependence of dense water accumulation and flushing on the variable Atlantic inflow, and (7) a robust, tidally driven circulation on the Svalbardbanken and around Bear Island. The Polar Front along the Svalbardbanken is fairly stationary, although its location is highly variable in the Sentralbanken area as a result of underflows (and winds—which were not simulated). The residence time for the Arctic Surface Water on Sentralbanken is about 8 months. Comparisons with available field measurements show a validation that is better than existing numerical model simulations. Entrainment of Arctic Surface Water on Svalbardbanken to the Atlantic inflow holds the Polar Front sharp and modifies the Atlantic Water as it flows to the Arctic Ocean. The simulated warm-core jet along this slope had a core speed up to 85 cm s −1 , whereas the best available current measurements near the core show surges up to about 30 cm s −1 . The simulated vorticity of the current is −0.33f, where f is the planetary vorticity. This can be provided from the conservation of potential vorticity. Both field data and laboratory simulations show that particles trapped in the Bear Island Current take 5–8 days to circle the island, which is 20 km in diameter. Except for surface confetti, agreement between model and field data was good for the southern flow east of Sentralbanken, but poor for the Murman Current. A model ‘wind’ caused a significant departure in this region and may be responsible for an exaggerated warm-core jet past Svalbardbanken.

Laboratory and numerical simulation of the Skagerrak circulation

Abstract A rotating laboratory model and two, three-dimensional numerical hydrodynamic models are used to simulate the stationary circulation of the Skagerrak region. The laboratory experiment is used as a reference solution (benchmark) for numerical simulations with the same set-up and forcing, except for some minor details. The laboratory model is validated to field data for 3 different processes: the Atlantic Water circulation along the bottom slopes, the Skagen front at the mouth of the Kattegat and deep-water renewal in Oslofjorden. The benchmark includes a map of the surface currents, several current velocity sections, salinity distributions and the response of the water column outside Oslofjorden to the spreading and dilution of tracers discharged in the German Bight and the Kattegat. Wind forcing and time-dependent flows like tides and frontal instabilities are not included in the benchmark. Local wind forcing is not simulated; however, the advective boundary conditions are a result of remote, regional meteorological forcing. The main validation to field data is for the SKAGEX experiment, during which time the local wind forcing was weak. Two numerical models are used in this study. A non-hydrostatic model MIKE 3 that employs an artificial compressibility technique and a hydrostatic model, the Princeton Ocean Model (POM), which uses sigma coordinates in the vertical. The results show that both models are capable of simulating large-scale geophysical flows in a region with steep topography and large density gradients. In the former case, improved results are obtained when the artificial compressibility is chosen as a calibration parameter. In the latter case a z-level interpolation method is used to help minimize potential pressure gradient errors and reasonable results are obtained with this model. However, the different response of the two numerical models highlights the need for model improvements in order to enable the simulation of the strong horizontal current shears in these waters. The results show the importance of potential vorticity dynamics in steering the slope currents and the baroclinic jets along the coasts.

Flow Around the Free Bottom of Fish Cages in a Uniform Flow With and Without Fouling

This paper explores the flow around fish cages in a uniform flow with the focus on the flow patterns close to the bottom of the models. Towing tests were conducted with six straight cylinders with the porosities 0%, 30%, 60%, 75%, 82%, and 90%, two cylinders with an inclination of 12.5 deg, and the porosities 0% and 75% and two cylinders with an inclination of 25 deg and the porosities 0% and 75%. The models all had a height-to-diameter ratio of 3 and were made from metal mesh. The Reynolds number was 5000 based on the diameter of the models and 15 based on the diameter of individual strings of the mesh for all tests. Particle image velocimetry, a nonintrusive optical technique, was used to analyze the flow around the models in the plane of symmetry through the center of the cylinders. The porosities of 82%, 75%, and 60% correspond to those of a clean fish cage netting in Norwegian Salmon farming with no fouling, light fouling, and heavy fouling, respectively. The inclinations of 12.5 deg and 25 deg reflect the inclination of the net of a commercial fish cage in a slow and a fast current, respectively. The Reynolds number of the strings was within the range of Reynolds numbers occurring on fish cages along the Norwegian coast. The results from this study are discussed with respect to the flow around and through the same models at identical Reynolds numbers. It is shown that the inclination of the net cage and fouling of the netting have major effects on the flow pattern around fish cages. The flow around and through net cages defines the water exchange within fish cages and the distribution patterns of particles and nutrients released from a net-pen. The information provided in this study can be valuable for the fish farming industry, as the decrease of the porosity due to fouling, as well as the deformation of the netting of fish cages, can be controlled by fish farmers.

Monitoring the Faroe Current using altimetry and coastal sea-level data

Abstract The possibility of using a coastal sea-level measurement program to compute flows in the Faroe Current (FC), north of the Faroe Islands, is studied using 3 years of Acoustic Doppler Current Profiler data together with water-level data at Torshavn in the Faroes, and altimetry data from a region to the north of the islands. A significant correlation is found between the sea-level rise across the FC and both the flow of Atlantic water and the total flow. Based on this correlation, linear algorithms are suggested between the surface slope and the flows. A seasonality is found in open-ocean sea-level from altimetry, steric heights from hydrography and the coastal sea-level, all having a minimum in early March when the observed inflows are at a maximum.

The Effects of Fish Cages on Ambient Currents

Experiments were carried out to measure forces on and wake characteristics downstream from fish cages. Cylinders made from metal mesh with porosities of 0%, 30%, 60%, 75%, 82%, and 90% were tested in a towing tank. The drag force was measured with strain gauges, and the flow field downstream from the models was analyzed using particle image velocimetry. The Reynolds numbers ranged from 1000–20,000 based on the model diameter and 15–300 based on the diameter of the strings of the mesh as an independent obstacle. High porosities (here, 82% and 90%) lead to low water blockage and allow a substantial amount of water to flow through the model. The data indicate that the wake characteristics change toward the wake characteristics of a solid cylinder at a porosity just below 75%. The drag force is highly dependent on the porosity for high porosities of a cylinder.

Pneumatic oil barriers: The promise of area bubble plumes

Reviews of bubble curtain oil herding studies in 1971 and in 1997 concluded that a bubble oil boom, or pneumatic oil barrier, is ineffective for retaining oil spills except in quiescent water, such as harbors. A bubble oil boom generates a sea-surface outwelling flow that traps or blocks oil. The primary bubble oil boom failure mode arises from oil droplet injection due to turbulence and instabilities at the oil slick front, where the outwelling flow balances the oil spreading. Bubble oil boom leakage occurs where these droplets are entrained into and pass through the bubble barrier. Increasing bubble flow creates stronger outwelling flows but increases turbulence and instabilities, leading to enhanced oil droplet entrainment. Natural seep observations, field trials, and laboratory studies demonstrate that a bubble plume with a wide bubble oil boom area, which is driven by an array of several parallel spargers (a bubble raft), can increase oil retention greatly while addressing key bubble oil boom failure modes compared with a line-source bubble curtain plume. Further improvements are identified by synergistic bubble oil boom application with a retaining skirt, dramatically improving the bubble oil boom performance. Specifically, the bubble oil boom keeps the oil distant from the skirt, minimizing or eliminating several conventional oil boom failure mechanisms. Also, entrained droplets, which easily traverse a single bubble curtain, are blocked effectively by a wide bubble plume curtain.

Average Flow Inside and Around Fish Cages With and Without Fouling in a Uniform Flow

The average flow field inside and around the bottom of porous cylinders in a uniform flow is explored using Particle Image Velocimetry (PIV). Tests were conducted on six cylinders with porosities of 0%, 30%, 60%, 75%, 82% and 90% in a flume tank where the flow field inside and around the models is time averaged over 180 seconds. The models had a height-to-diameter ratio of 3 and were made from metal mesh. The Reynolds numbers ranged from 5,000 to 20,000 based on the diameter of the models and from 75 to 300 based on the diameter of individual strands of the mesh, which corresponds to the Reynolds numbers occurring at salmon fish cage netting used along the Norwegian coast. The porosities of 82%, 75% and 60% correspond to those of a fish cage netting in Norwegian Salmon farming with no, light and heavy biofouling, respectively. The results from this study are discussed with respect to the instantaneous flow field in and around the same cylinders at identical Reynolds numbers. The focus is on the effect of porosity on the ventilation inside the cages and the vertical transports within the near wake. It is shown that heavy fouling of aquacultural nettings can lead to internal circulation inside fish cages and therefore has the potential to reduce the ventilation of the net pens dramatically. The description of the time-averaged flow field inside and around porous cylinders can be used as benchmarks to validate and adjust numerical models of the flow past porous cylinders. The results from this study can be valuable also for the fish farming industry, as bio-fouling and the reduced porosity of fish cages can be monitored and controlled directly by fish farmers.© 2010 ASME