WWW.WAS.ORG • WORLD AQUACULTURE • DECEMBER 2015 25 similarity cannot be achieved simultaneously. Furthermore, for flow around a fish cage structure in two dimensions, there are two characteristic length scales (cage diameter and twine diameter) that can be applied to Reynolds or Froude number scaling. Since dynamic similarity cannot be achieved for these flow conditions, achieving geometric similarity is of primary importance. A scale model has geometric similarity if all dimensions have been scaled appropriately and the model maintains the same overall shape as the full-scale application. For a flexible body such as fish cage netting, it is important to match the overall shape of the full-scale body. To achieve this, the ratio of the global horizontal to vertical forces acting on the netting between full scale and model scale must be matched. The only horizontal force acting on the netting is hydrodynamic drag, while vertical forces include hydrodynamic lift, apparent weight in water of the netting, and added weight on the netting. The amount of weight hung from cages and scale model testing velocity could be adjusted to maintain the ratio of horizontal to vertical forces, thus achieving the proper deformed cage shape and geometric similarity (Turner et al. 2015). The 1:15 scale model circular fish cages were designed from those used in commercial salmon farming, with high detail (Fig. 2). Cages were made of two concentric floating hoops, with attached netting hanging below the waterline. A predator net was attached to an outside hoop and the inner containment net was attached to an inside hoop, with both connected to the bottom weight collar. The model cage had an outside diameter of 2.12 m and an inside diameter of 1.97 m. The weight collar at the cage bottom was used to maintain the cylindrical cage shape when the netting experienced drag forces from currents. Chain link was hung from the weight collar for additional weight. The amount of weight that can be added depends on the amount of flotation the cages offer. The amount of weight used also influences the test current velocity, as scaling laws must be maintained, ensuring the overall shape of the cage is matched between full scale and model scale. Knotless nylon netting for the scale model cages was supplied and fully constructed by GMG Fish Services Ltd., a division of Cooke Aquaculture, Inc. Porosity of the scale-model netting was 0.7 for the containment net and 0.83 for the predator net. This influence current flow on the scale of meters is important for IMTA site design and estimates of oxygen supply and nutrient uptake. Near-field hydrodynamics is also of relevance to the attachment potential of some ectoparasites such as sea lice (Chang et al. 2011) and consequent fate of therapeutants used for in situ cage treatments (Ernst et al. 2014). Sea lice attachment is less prevalent under conditions of high current velocities (Revie et al. 2003), suggesting location of fish cages in more exposed environments could reduce infection potential. However, the aforementioned effects of cages and array configuration have the potential to reduce current flow, potentially negating advantages of location in high-energy environments. Bath treatments for sea lice may be administered in well-boats or temporarily tarped cages and therapeutants will be ultimately released into the water column after treatment (Ernst et al. 2014). While toxicity potential to near-field, co-cultured and wild aquatic animals is a function of the therapeutant used and its half-life, it is also a function of concentration and exposure duration (Burridge 2013) and therefore current flow. This study investigated current dynamics of water flow through a 1:15 model circular cage-array deployed in a flume tank. The approach to this study was unique in that data was collected over an intensive three-day graduate student workshop, consisting of two mechanical engineering, three geography, one oceanography and two biology students (Fig. 1). The training objective was to apply a team-based research approach to a complex problem, under conditions of time and resource limitations, with a publishable research document as a deliverable. The research objective was to document mixing behavior and wake morphology of current flow passing through a model cage array to guide the optimal placement of co-culture species at IMTA fish farms. Study Methods Model scaling, construction and deployment. For the fish cage model to be properly scaled, it must be geometrically and dynamically similar to the full-scale cage. Dynamic similarity is satisfied if the full scale and model scale have identical fluid flows. Two important similarity parameters to consider are the Reynolds number (), the ratio of inertial forces to viscous forces, and the Froude number (), the ratio of inertial forces to gravity forces. When testing in the same fluid, Reynolds and Froude (CONTINUED ON PAGE 26) TOP, FIGURE 3. Scale (1:15) model cage and array. Cage array shown deployed in the flume tank in a 2 x 2 configuration. BOTTOM, FIGURE 4. Neutrally buoyant drogue ball in transit.
RkJQdWJsaXNoZXIy MjExNDY=