WO2009085987A2

WO2009085987A2 - Apparatus and method for offshore ocean cage aquaculture

Info

Publication number: WO2009085987A2
Application number: PCT/US2008/087487
Authority: WO
Inventors: Homer Rudolph Schmittou; Michael Craig Cremer; Cliff Goudey; Thomas Schmittou; Jesse Chappell
Original assignee: United States Soybean Export Council
Priority date: 2007-12-19
Filing date: 2008-12-18
Publication date: 2009-07-09
Also published as: WO2009085987A3

Abstract

A mooring system for a fish cage used for aquaculture includes a first rigid tubular member. An anchor arranged at seabed is coupled to the first rigid tubular member via a mooring chain. The mooring chain defines a moving radius of the floating element. A buoyancy of the fish cage is adjustable by varying the amount of fluid located in the first rigid tubular member. The buoyancy of the fish cage is adjusted to allow the fish cage to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

Description

APPARATUS AND METHOD FOR OFFSHORE OCEAN CAGE AQUACULTURE

Background

[0001] Unlike fishing, aquaculture, also known as aquafarming, implies the cultivation of aquatic populations under controlled conditions. A principal method of aquaculture is fish farming, which involves raising fish commercially in tanks or enclosures. A facility that releases juvenile fish into the wild for recreational fishing or to supplement a species' natural numbers is generally referred to as a fish hatchery. Fish species raised by fish farms may include salmon, catfish, tilapia, cod, carp, trout and others.

[0002] Aquaculture has traditionally been a labor and cost intensive industry. One such cost of aquaculture is associated with net or net cages conventionally used to raise the fish. Conventionally large circular net cages have been used for offshore production of fish. These cages typically float at the surface of the ocean and may be damaged or destroyed by inclement weather, such as Typhoons and Hurricanes. Therefore, there remains a need in the art for an apparatus and method for offshore ocean cage aquaculture that overcomes the problems associated with the prior art.

[0003] Another cost is associated with the feed used to raise the fish. Basically there are two kinds of aquaculture: extensive aquaculture, which is based on local food supply, and intensive aquaculture, in which the fish are fed with external food supply. Conventionally in intensive aquaculture, "trash fish" (noncommercial fish) have been used as feed. The use of trash fish as feed, however, can also contribute to poor water quality and high rates of disease outbreak. Thus, other forms of feed may be desirable for operating a more efficient aquaculture system.

Summary

[0004] According to embodiments of the present invention, there is provided a more efficient and cost-effective aquaculture system and method. The system of the invention may include a fish cage attached to an anchor that limits the dislocation of the fish cage within the water. The buoyancy of the fish cage may be controllable via a ballast such as a chain attached at the bottom of the cage, as well as water ballast flowing through the frame of the fish cage. The buoyancy of the fish cage may be managed such that with increased water current resulting from a storm or a typhoon, the fish cage submerges in the water, thus preventing the fish cage and the fish from being damaged by the typhoon. In addition, the cage provides a secure feeding mechanism to provide feed, which may be soy-product based, to the fish without risking the loss of the feed under various operating conditions.

[0005] According to an exemplary embodiment of the invention, there is provided a fish cage, including a frame comprising at least a first rigid tubular frame member; a port located on the first tubular member, the port adapted to seal a fluid within the first tubular member; and a non-collapsing containment net supported by the frame. The buoyancy of the frame is adjustable by varying the amount of fluid located in the first tubular member. The fish cage is coupled to an anchor at seabed, the coupling to the anchor defining a moving radius within which the fish cage is contained. The buoyancy of the frame is adjusted to allow the fish cage to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

[0006] In a further exemplary embodiment, the fish cage may include a feed containment area including a plurality of splashboards and a feed containment net containing the feed, the feed containment area being mounted on top of the frame. The fish cage may also include a ballast attached to the frame, the ballast being adapted to offset at least a portion of the buoyancy of the frame. A netting panel of the containment net may include a sealable gap, the gap being openable and connectable to a tunnel through which the fish are transferred to a second fish cage. In an exemplary embodiment, the frame of the fish cage may be shaped as a truncated pyramid.

[0007] In an exemplary embodiment, the first rigid tubular member may include at least one valve for allowing fluid in or out of the first rigid tubular member. In a further embodiment, the first rigid tubular member may include a first valve arranged near an upper portion of the first rigid tubular member that allows the fluid to flow in or out of the first rigid tubular member; a hose coupled to the first valve on one end that stretches near a lower portion of the first rigid tubular member; and a second valve arranged near the upper portion of the first rigid tubular member that allows air to move in or out of the first rigid tubular member, wherein the amount of fluid located in the first rigid tubular member is varied by adding or removing fluid through the first valve. [0008] According to another aspect of the invention, there is provided a single point mooring system, which includes an anchor arranged at seabed; a mooring chain coupled to the anchor at a first end; and a floating element including a first rigid tubular member coupled to a second end of the mooring chain, wherein a buoyancy of the floating element is adjustable by varying the amount of fluid located in the first rigid tubular member, the mooring chain defining a moving radius of the floating element, wherein the buoyancy of the floating element is adjusted to allow the floating element to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

[0009] In a further embodiment, the single point mooring system may include at least one bridle coupling the floating element to the second end of the mooring chain. The mooring system may also include a connector coupling the at least one bridle to the second end of the mooring chain. The mooring system may also include a buoy coupled to the connector, wherein buoy being adapted to offset at least the weight of the connector. Further, there may be included a ballast coupled to the floating element to offset at least a portion of the buoyancy of the floating element. An upper portion of the chain is lighter relative to a lower portion of the chain. [00010] According to a further embodiment, the floating element may be a fish cage including a frame shaped as a truncated pyramid, a containment net positioned inside the frame, and a feed containment area positioned on top of the frame, wherein the first rigid tubular member includes a sloping member of the frame.

[00011] In a further embodiment, there may be provided at least one quick release buoy attachable to the floating element, the buoyancy of the floating element being adjusted such that the floating element submerges in the water substantially close to the water surface when the at least one quick release buoy is detached from the floating element, but float at least partially on water when the at least one quick release buoy is attached to the floating element. [00012] In an exemplary embodiment, the first rigid tubular member of the mooring system may include a first valve arranged near an upper portion of the first rigid tubular member that allows the fluid to flow in or out of the first rigid tubular member; a hose coupled to the first valve on one end that stretches near a lower portion of the first rigid tubular member; and a second valve arranged near the upper portion of the first rigid tubular member that allows air to move in or out of the first rigid tubular member, wherein the amount of fluid located in the first rigid tubular member is varied by adding or removing fluid through the first valve. [00013] According to another aspect of the invention, there is provided a mooring method, including: providing a fish cage including a frame comprising at least one rigid tubular frame member; and adding a liquid into the at least one rigid tubular frame member through a port, or removing a liquid from the at least one rigid tubular frame member through a port, thereby adjusting the buoyancy of the fish cage; wherein the fish cage is coupled to an anchor at seabed, the coupling to the anchor defining a moving radius within which the fish cage is contained, wherein the buoyancy of the frame is adjusted to allow the fish cage to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

[00014] According to a further embodiment of the invention, the mooring method may include: providing an anchor on seabed; providing a mooring chain coupled at a first end to the anchor; providing a connector coupled to a second end of the mooring chain; providing at least one bridle coupling the fish cage to the connector; and providing a buoy coupled to the connector to offset at least the weight of the connector. The mooring method may also include providing at least one quick release buoy attachable to the fish cage, the buoyancy of the fish cage being adjusted such that the fish cage submerges in the water substantially close to the water surface when the at least one quick release buoy is detached from the floating element, but floats at least partially on the water when the at least one quick release buoy is attached to the fish cage.

Brief Description of the Drawings

[00015] The foregoing and other features of the invention will be apparent from the following, more particular description of exemplary embodiments of the invention, as illustrated in the accompanying drawings wherein like reference numbers generally indicate identical, functionally similar, and/or structurally similar elements. The left most digits in the corresponding reference number indicate the drawing in which an element first appears.

[00016] FIG. 1 is a perspective view of an exemplary fish cage according to the present invention;

[00017] FIG. 2 is a perspective view of an exemplary feed containment area of a fish cage according to the present invention;

[00018] FIG. 3 is a partial, cross- sectional view of an alternative embodiment of a feed containment area according to the present invention; [00019] FIG. 4 is a perspective view of another alternative embodiment of a feed containment area according to the present invention;

[00020] FIG. 5 is a top view of an exemplary work platform for a fish cage according to the present invention;

[00021] FIG. 6A is a perspective view of an exemplary netting of a fish cage according to the present invention;

[00022] FIG. 6B is a view of an exemplary side panel of a containment net of a fish cage according to the present invention;

[00023] FIG. 6C is a partial, detail view of the exemplary feed containment net of FIG.

6A;

[00024] FIG. 6D is a perspective view of an exemplary feed containment net placed inside a feed containment area according to an embodiment of the present invention;

[00025] FIG. 6E is a perspective view of an exemplary feed containment net placed inside a containment net, according to an alternative embodiment of the present invention;

[00026] FIG. 7 is a side view of an exemplary mooring system according to the present invention;

[00027] FIG. 8A is a side view and FIG. 8B is a top view of an exemplary anchor for use with the mooring system of FIG. 7;

[00028] FIG. 9A is a top view of an exemplary mooring connector for use with the mooring system of FIG. 7;

[00029] FIG. 9B is a schematic representation of a fish cage attached to the mooring connector of FIG. 9A;

[00030] FIG. 1OA is a schematic representation of an exemplary water ballast for a fish cage according to the present invention;

[00031] FIG. 1OB is a schematic representation of an alternative embodiment of a water ballast for a fish cage according to the present invention;

[00032] FIG. HA is a perspective view of an exemplary fish cage with quick release buoys according to the present invention;

[00033] FIG. HB is a perspective view of the exemplary fish cage of FIG. HA, shown after detachment of the quick release buoys, and upon partial sinking of the fish cage;

[00034] FIG. HC is a perspective view of the exemplary fish cage of FIG. HA, shown after detachment of the quick release buoys; [00035] FIG. HD is a perspective view of the exemplary fish cage of FIG. HC, shown upon further sinking of the fish cage; and

[00036] FIG. 12 is a perspective view of an exemplary fish transfer mechanism according to the present invention.

Detailed Description

[00037] Exemplary embodiments of the invention are discussed in detail below. While specific exemplary embodiments are discussed, it should be understood that this is done for illustration purposes only. In describing and illustrating the exemplary embodiments, specific terminology is employed for the sake of clarity. However, the invention is not intended to be limited to the specific terminology so selected. A person skilled in the relevant art will recognize that other components and configurations may be used without departing from the spirit and scope of the invention. It is to be understood that each specific element includes all technical equivalents that operate in a similar manner to accomplish a similar purpose. Each reference cited herein is incorporated by reference. The examples and embodiments described herein are non-limiting examples.

[00038] According to an exemplary embodiment of the invention, a submersible offshore finfish aquaculture system may comprise a fish cage attached to an anchor such as a single point anchor, which may allow the cage to float down current from the anchor and to submerge automatically in response to increased, storm- generated, wind and/or water currents. By submerging during a storm, the cage, and the fish inside, can be protected from the resulting high wave conditions.

[00039] Referring to FIG. 1, there is depicted an exemplary sectional view of a fish cage

100, according to an exemplary embodiment of the invention. According to this embodiment, the fish cage 100 may include a frame 102 shaped, for example, in the shape of a truncated pyramid. The frame may include a top portion 104 having four upper horizontal members 114a- 114d and a bottom portion 106 having four lower horizontal members 116a- 116d. Four sloping members 118a-118d can attach the four corners 124a- 124d of the top portion 104 to the four corners 126a- 126d of the bottom portion 106, respectively. The area defined by the bottom portion 106 may be substantially larger than the area defined by the top portion 104, although other configurations are possible. The sloping members 118a- l 18d may be arranged diagonally, and may each have their end join with the respective corner 126a- 126d at an acute angle with respect to the plane defined by the bottom portion 106. This angle may be, e.g., between about 40 ° and about 55 °. At the upper end, the upper ends of the sloping members 118a-118d may join the respective corners 124a- 124d at an obtuse angle with respect to the plane of the top portion. This angle may be, e.g., between about 125° and about 140°. One of ordinary skill in the art will appreciate that the fish cage 100 is not limited to the truncated pyramidal shape shown, and that other shapes, such as box-shaped, cone-shaped, etc., are contemplated by the present invention.

[00040] According to an exemplary embodiment of the invention, the upper horizontal members 114a- 114d, the lower horizontal members 116a-116d, and the sloping members 118a- 118d may be constructed from rigid high-density polyethylene pipes, although other materials, such as polyvinylchloride (PVC), aluminum, or steel, can alternatively be used. In an exemplary embodiment, the pipes used for the upper horizontal members 114a-114d and the lower horizontal members 116a-116d may be approximately 280mm in diameter, and the pipes used for the sloping members 118a-l 18d may be approximately 200 mm in diameter. The pipes may be connected at the upper corners 124a- 124d and the lower corners 126a-126d via, for example, galvanized steel joints, however, other connectors known in the art can alternatively be used. [00041] In an exemplary embodiment, the fish cage 100 may be designed to accommodate a large number of fish. In an exemplary embodiment, the upper portion 104 may be, for example, 2m x 2m, and the lower portion 106 may be, for example, 7m x 7m. The height of the frame 102 may be, for example, 4.5m. However, the size of the frame 102 may be smaller or larger depending on the type and/or amount of fish intended to be contained in the cage 100. [00042] In an exemplary embodiment, the frame 102 may support a containment net 130, shown, for example, in FIG. 6A. The containment net 130 may be used to contain the fish located in the cage 100. Referring to FIG. 6A, the containment net 130 may have a shape substantially matching that of the respective frame 102. For example, the net 130 for use with the truncated-pyramidal-shaped frame 102 can comprise an assembly of five netting panels including four side panels 138a-138d and a bottom panel 134. The bottom panel 134 may be in the shape of a square and may attach to the four lower corners 126a-126d of the frame 102. Each of the four side panels 138a-138d may be in the shape of an isosceles trapezoid, sized to the frame 102. The bottom panel 134 and the size panels 138a- 138d can be attached to one another via a number of framing lines, described in detail later. [00043] According to an exemplary embodiment of the invention, the fish cage 100 may also include a feed containment area 140 arranged on the top portion 104 of the frame 102, as shown in FIG. 1. The feed containment area 140 may contain the feeding material, e.g. soy- based feeds or other feed pellets, used to feed the fish. The feeding material can be contained within one or more nets, placed on top portion 104 of the frame 102, as will be later discussed. The feed containment area 140 may include splashboards, such as the four splashboards 142a- 142d, respectively attached to the upper horizontal members 114a-114d. The splashboards 142a-142d may be welded, or otherwise attached, to the frame 102 to support the fish cage 100 against wave or current action.

[00044] FIG. 2 depicts an exemplary feed containment area 140 according to an exemplary embodiment of the invention. In this embodiment, the four upper splashboards 142a- 142d can be attached to the upper side of respective pipes of the upper horizontal members 114a-114d. There can also be four lower splashboards 144a- 144b attached to the lower side of respective pipes of the upper horizontal members 114a- 114d. Each upper splashboard 142a- 142d may be supported by a plurality of upper standpipes 146a- 146c, attached vertically to each of the upper horizontal members 114a- 114d, although other structures for supporting the upper splashboards are possible. Each lower splashboard 144a- 144d may be supported by a plurality of lower standpipes 148a- 148c, arranged opposite the upper standpipes 146a- 146c, also attached vertically to each of the upper horizontal member 114a-114d, although other structures for supporting the lower splashboards are possible. The upper standpipes 146a- 146c and the lower standpipes 148a- 148c may be welded, bolted, riveted, bonded, or otherwise attached to each of the upper and lower splashboards, as well as the upper horizontal members 114a- 114d. [00045] According to an exemplary embodiment, the upper splashboards 142a-142d and the lower splashboards 144a- 144d may be made of polyethylene, high density polyethylene, PVC, steel, aluminum, or other materials known in the art. The upper standpipes 146a- 146c and lower standpipes 146a-146c may also be made of polyethylene, high density polyethylene, PVC, steel, aluminum, or other materials known in the art. The upper splashboards 142a-142d may be initially installed with voids between the adjacent upper splashboards, and the lower splashboards 144a- 144d may likewise be initially installed with voids between them. After the splashboards are installed, the voids between adjacent upper splashboards 142a- 142d can be filled with upper curved fillers 152a-152d, which may be screwed, welded, riveted, or otherwise attached to the adjacent sides of the upper splashboards 142a- 142d. Similarly, the voids between adjacent lower splashboards 144a-144d can be filled with lower curved fillers 154a- 154d. The fillers may be made out of plastic for flexibility and easy installation, although other materials, such as sheet metal, plastics, and composites can also be used.

[00046] According to an alternative embodiment of the invention depicted in the cross- sectional view of FIG. 3, the upper splashboards 142a-142d and/or the lower splashboards 144a- 144d may be attached to the respective upper horizontal member 114a- 114d via one or more gussets, for example, the upper gussets 324 and the lower gussets 326, respectively, shown in FIG. 3. The gussets 324, 326 can be substantially triangular shaped and can support the splashboards against wave action, although other shapes are possible. In an exemplary embodiment, three spaced-apart gussets 324, 326 can be provided for each splashboard, attached on one side to the splashboard and to the respective upper horizontal member 114a-114d on the other. As illustrated in FIG. 3, an upper splashboard 144a may be positioned at an acute angle 308 with respect to the projection 304 of the diameter 302 of the upper horizontal member 114a (i.e. , line 304 is normal to the circumference of the horizontal member 114a). The acute angle may be, e.g., about a 27° angle. According to this exemplary configuration, the upper splashboard 142a may form an approximately 18° angle with respect to a vertical projection 306 of the upper horizontal member 114a, as shown in FIG. 3. The lower splashboard 144a may additionally or alternatively be positioned at an acute angle 318, e.g., about 27°, with respect to a projection 314 of a diameter 312 of the cross-section of the upper horizontal member 114a, as shown in FIG. 3.

[00047] According to a further exemplary embodiment of the invention as depicted in

FIG. 4, in addition to the splashboards, the feed containment area 140 may also comprise one or more rails 402a-402c, extending vertically upward from each of the upper horizontal members 114a-l 14d. Each rail 402a-402c can be attached to one or more vertical support elements 404a- 404c respectively mounted on top of or adjacent to the upper splashboards 142a- 142d. Alternatively, the horizontal support elements 404a-404c may be directly mounted on top of the upper horizontal member 114a- 114d. In yet another embodiment, the horizontal support elements 404a-404c may be extensions of the corresponding upper standpipes 146a- 146c. In addition, each rail 402a-402d may also include one or more horizontal support elements 406. The rails 402a-402c can assist the personnel attending the fish cage 100 in mounting the feed containment area 140 by standing on an upper horizontal member 114a- 114d and grabbing the horizontal support element 406. [00048] According to an alternative embodiment, the feed containment area 140 may comprise one or more work platforms 502, which can be attached horizontally, for example, to one or more of the upper horizontal members 114a- 114d. FIG. 5 depicts a top view of a work platform 502 including a flat plate that can be mounted to one of the upper horizontal members, for example, upper horizontal member 114a. According to an exemplary embodiment, a work platform 502 can be mounted to each of the upper horizontal members 1 14a-114d, providing a work platform that extends substantially around the outer periphery of the feed containment area 140. According to an exemplary embodiment, each work platform 502 can be wide enough to support one or more personnel standing on it, and can provide access to the feed containment area 140 in order to perform fish husbandry activities. Each platform 502 may be bolted to the respective upper horizontal member 114a- 114d using, for example, a plurality of U-bolts extending through holes located in the upper horizontal members 114a- 114d, however, other fabrication and attachment techniques known in the art can be used to provide the work platform(s) 502. One of ordinary skill in the art will appreciate that a single, continuous work platform 502 can be provided in place of the multiple, partial work platforms 502 shown in FIG. 5.

[00049] FIG. 6A depicts an exemplary netting according to an exemplary embodiment of the invention. The netting may include a containment net 130, which may have a shape that generally corresponds to the frame that it will be used in conjunction with. For example, when used with the truncated-pyramidal cage 102 of FIG. 1, the containment net 130 may include four side panels 138a-138d, and a bottom panel 134; the bottom panel 134 may be in the shape of a square and may attach to the four lower corners 126a- 126d of the frame 102. Each of the four side panels 138a-138d may be in the shape of an isosceles trapezoid, approximately sized to the frame 102.

[00050] In an exemplary embodiment, in addition to the containment net 130, the netting may include a top net 160 and a feed containment net 162. The top net 160 may be designed to approximately fit the upper portion of the containment net 130. In an exemplary embodiment, after the fish are placed inside the containment net 130, the top net 160 may be placed on and tied to the upper portion of the containment net 130 to seal the fish inside the containment net 130. The feed containment net 162, in turn, may be located inside the top portion of containment net 130 and below top net 160, providing a containment for the feed. Alternatively, the feed containment net 162 may be an independent small mesh net that prevents feed escapement. Alternatively, the feed containment net 162 may be located inside the containment net 130, attached to one of the four side panels 138a-138d, or above the top net 160. [00051] Each panel of the containment net 130 may include a perimeter rope 602 defining the shape of the panel, and a netting 604. FIG. 6B shows an exemplary side panel 138a of the containment net 130, including the netting 604 and the perimeter rope 602. The perimeter rope 602 may take the shape of an isosceles trapezoid. The netting 604 may be connected to the perimeter rope 610 via a plurality of lashings 612, although other types of connections are possible.

[00052] Referring to FIG. 6C, a more detailed exemplary view of a lower corner of the containment net 130 is depicted according to an exemplary embodiment of the invention. In this view, the netting 604 may include knotted stretch mesh 606. The netting mesh 606 may be made of braided nylon twine and may be between about 2 and about 3 millimeters in diameter, however other materials and sizes are possible. According to an exemplary embodiment, the breaking strength of the nylon twine may be over 300 lbs. The netting mesh 606 may be knotted about 1 to 3 centimeters apart, forming squared-shape net mesh eyes 608 of approximately 1-9 sq. centimeters in area, however other configurations and sizes are possible, depending, for example, on the type of fish contained by the net, and the type of feed used. [00053] The netting mesh 606 can be edged with a perimeter rope 610, which can help maintain the shape of the panel and can facilitate individual replacement of the panels. The perimeter rope 610 may also be made of braided nylon, but may be thicker than the netting mesh 606, e.g., 6- 10 mm in diameter, although other materials and dimensions are possible. The dimensions of the perimeter rope 610 for each side panel 138a- 138d and the bottom panel 134 of the containment net 130 can substantially correspond to the shape and size of the frame 102. According to an exemplary embodiment, dimensions of the containment net 130 may be slightly smaller than the dimensions of the frame 102, in order to provide a tighter fit of the panels on the frame 102, for example, in order to reduce movement of the containment net 102 with respect to the frame 102 under wavy conditions.

[00054] Still referring to FIG. 6C, the side panels 138a- 138d and the bottom panel 134 may be coupled to one another via a plurality of framing ropes 614, for example, to form the truncated-pyramidal shape shown in FIG. 6A. The framing ropes 614 may define the outer perimeter of the bottom portion and the sides of the containment net 130. The framing ropes 614 may be thicker than the perimeter ropes 610, e.g., about 15-20 mm in diameter, although other configurations are possible. The framing ropes 614 may be made of braided polyester, although other materials are possible. Each framing rope 614 may include an eye splice 618 at each end to facilitate interconnection between the framing ropes 614, as depicted in the exemplary view of FIG. 6B. The eye splices 618 may each include an aperture that may be, for example, 10 cm in width. The perimeter lines of the side panels 138a- 138d and the bottom panel 134 may be attached to the framing ropes 614 via a plurality of seizings 616. The framing ropes 614 may be shackled together to facilitate attachment of the containment net 130 to the frame 102, for example, at its lower corners 126a-d. The framing ropes 614 on the upper portion of the containment net 130 may be attached to the upper corners 124a-124d of the frame, or to the splashboards 144a- 144d, although other attachment points are possible. [00055] FIG. 6D depicts an exemplary view of a feed containment net 162 placed inside the feed containment area 140, according to an exemplary embodiment of the invention. In an exemplary embodiment, the feed may be sandwiched by the top net 160 and the feed containment net 162. The feed containment net 162 may be larger than the top net 160 such that the edges of the feed containment net 162 attach to the upper portions of the feed containment area 140, e.g., the top of the upper splashboards 142a-142d. In an exemplary embodiment, the four corners 164a-d of the feed containment area 164a- 164d may stretch over the four filler 152a-152d and tie to the four corners 124a-124d of the top portion 104 of the frame 102. In an alternative embodiment, the upper edge of the upper splashboards 142a- 142d may be provided with small hopes (not shown), to which the feed containment net 162 may be attached via a plurality of lashings. In an exemplary embodiment, since the feed containment net 162 is larger than the are of the top net 160, it is flexible to accommodate a large amount of feed being contained between the top net 160 and the feed containment net 162.

[00056] FIG. 6E depicts an exemplary view of the feed containment net 162, arranged inside the fish containment net 130, according to an alternative embodiment of the invention. As shown in FIG. 6E, the feed containment net 162 may include a mesh feed net placed underneath the top net 160 around the inside perimeter of the upper corners of the top net 160, to contain the floating feed pellets. Alternatively, the feed containment net 162 may be attached to the inside perimeter of an upper portion of the fish containment net 130, below the top net 160. In an exemplary embodiment, the fish containment net 162 may be hanging from the fish containment net 162 below its attachment points. After deployment of fish 100 in the water, the feed containment net 162 may be extended upward and attached, at various points, to the top net 160, thus securing the feed contained in the fish containment net 162.

[00057] Referring now to FIG. 7, an exemplary mooring system and method is now discussed according to an exemplary embodiment of the invention. The fish cage 100, discussed above, may have sufficient buoyancy to stay afloat on the water. To prevent waves or currents from carrying the fish cage 100 away, the cage may be held in place via a single point mooring system. The exemplary single point mooring system depicted in FIG. 7 may comprise an anchor 704, coupled to the fish cage 100, and positioned on the seabed. The anchor 704 is described in detail with reference to FIGs. 8A-B. The anchor 704 may be coupled to the fish cage 100, for example, via a mooring chain 702. In an exemplary embodiment, the mooring chain 702 is coupled to the fish cage 100 via a mooring connector 706, in which case the mooring chain 702 is connected at one end to the anchor 704 and at the other end to a mooring connector 706. The mooring connector 706 is described in detail later with reference to FIGs. 9A-B. The fish cage 100 may be connected to the mooring connector 706 via one or more bridles 710. Also connected to the mooring connector 706 may be a buoy 708, which may include sufficient buoyancy to offset the weight of the mooring connector 706 and the chain 702. [00058] The mooring chain 702 may include two sections - a lower section 702a that may be made of a heavy chain that typically rests on the seabed adjacent to the anchor 704, and an upper section 702b that may be made of a lighter chain, allowing it to float in the water along with the fish cage 100. The lower section 702a may be connected to the upper section 702b via a swivel, or other connection device known in the art. In an exemplary embodiment, the lighter upper section 702b may stretch out horizontally during storm conditions, causing the fish cage 100 to submerge.

[00059] In an exemplary embodiment of the invention, the upper section 702a of the mooring chain 702 may lead to a mooring connector 706, which in turn may connect to the buoy 708 and the bridles 710. The buoy 708 may be capable of flotation on water and may have sufficient buoyancy to offset the weight of the upper section 702b of the mooring chain 702. However, the buoyancy of the buoy 708 may not be too much so as to prevent the fish cage 100 from submerging. Thus, in an exemplary embodiment, the buoyancy of the buoy 708 may be just enough to keep the upper section 702b of the mooring chain 702 afloat. In an exemplary embodiment, the buoy 708 may provide an additional advantage in that it may absorb current- induced loads on the mooring chain 702 before these loads impact the fish cage 100, thus preventing the fish cage 100 from getting damaged by the mooring chain 702. The buoy 708 may be a single unit made of steel, plastic or other material filled with gases such as air. The buoy 708 may also be made of Styrofoam or other materials with high buoyancy. Alternatively, the buoy 708 may be a complement of several individual buoys, each of which may be made of varying material. In an exemplary embodiment, the single buoy 708 or the complement of smaller buoys have a total displacement buoyancy sufficient to offset the weight of the upper section 702b of the mooring chain 702, but not so great as to prevent submersion of the buoy(s) 708 during storm events in conjunction with submersion of the fish culture cage. [00060] FIG. 8A depicts an anchor 704, according to an exemplary embodiment of the invention. The anchor 704 may include a block 802, formed, for example, from concrete, steel, lead, or other dense material. According to an exemplary embodiment, the block 802 may weigh over 10,000 pounds (e.g., 11,200 lbs), and may be in the shape of a truncated pyramid with sloping sides. This configuration may encourage embedding of the anchor 704 into the seabed, and may create more friction on the seabed, in order to provide higher resistance for the mooring. The four sides of the block 802 may slope at 45 degrees and may be truncated, for example, at about 1.22 sq. m. Anchor 704 may include reinforcements in order to prevent breakage in case of rough handling. For example, re-bar may be included in the concrete block 802 during the casting of the anchor 704. The re-bar, if included, may also be configured to provide a connection point for the mooring chain 702. According to the exemplary embodiment shown, the armature of the block 802 includes two layers of re-bar 804, 806 embedded within the block 802. One or both layers of the re-bar 804, 806 can be supported by and pass through the links of the mooring chain 702, as shown in FIG. 8A. According to an exemplary embodiment, the block 802 may be 2 meters tall, in which case, the lower layer 804 may be positioned approximately 10 cm from the bottom of the block 802, and the top layer 806 may be positioned approximately 20 cm from the top of the block 802. One of ordinary skill in the art will understand, however, that other shapes, sizes, and configurations of the anchor 704 are possible, including screw anchors, embedding rods or other anchor configurations that remain in a constant position on the seabed.

[00061] FIG. 8B depicts a top view of the block 802 according to an exemplary embodiment of the invention. Each layer of re -bar 804, 806 may include two re-bars 814a, 814b, 816a, 816b, arranged diagonally and passing through the chain links, and four re-bars 818a-818d, 820a-820d, arranged in a square at their outer ends, aligned parallel to the edges of the block 802. In this way, the re-bar can provide additional support for the mooring chain 702 inside the block 802. One of ordinary skill in the art will appreciate, however, that other configurations of the re -bar are possible. Alternatively, the block 802 can be constructed without the re-bar.

[00062] FIG. 9A depicts a mooring connector 706, according to an exemplary embodiment of the invention. The mooring connector 706 may be used to minimize contact between the bridals 710 and the mooring chain 702, which may undergo constant wave action. The mooring connector 706 may include five attachment points 902-910. The buoy 708 may be connected to the attachment point 902, for example, and the mooring chain 702 may be attached to the attachment point 904, for example.

[00063] FIG. 9B depicts the connection of a fish cage 100 to the mooring connector 706, according to an exemplary embodiment of the invention. In this embodiment, the attachment points 906-910 are provided for three bridals 710. In an exemplary embodiment, the lower bridals 710a, 710b may connect the two attachment points 908, 910 to the two lower points on the fish cage 100. The upper bridles 71Od, 71Oe, in turn, may attach to two points on an upper portion of the fish cage 100. The upper bridals 71Od, 71Oe, may converge to form a single bridle extension 710c that may lead to the attachment point 906 of the mooring connector 706. which are connected to an upper portion for the fish cage 100. The lower bridals 710a, 710b, and the two upper bridals 71Od, and 71Oe, attach to four points of the cage 100 to provide proper cage alignment in the water.

[00064] In an exemplary embodiment, the bridles may be polyester ropes, which may be, for example, 18-mm to 24-mm in diameter. The bridle ropes may be sized to meet specific peak loads, breaking strength, and tensions to provide a safe working load (SWL). In an exemplary embodiment, the maximum tensile strength (MTS) of the rope may be seven times the determined SWL. The bridle ropes 710 may have MTS ratings that, in an exemplary embodiment, vary from approximately 11,000 kg for the upper bridles 71Od, 71Oe, to 12,500 kg for the two lower bridles 710a, 710b and the upper extension bridle 710c.

[00065] According to an exemplary embodiment of the invention, it may be desirable for the fish cage 100 to submerge in the water in response to increasing water and wave current velocity. For example, during storms, typhoons, and hurricanes, as the water current picks up speed and pushes the fish cage 100 away from the anchor 104, it is desirable for the fish cage 100 to submerge in the water, thus decreasing the likelihood of the fish cage 100 getting damaged by the storm, typhoon, or hurricane. Therefore, the fish cage 100 may be adapted to have sufficient buoyancy to stay afloat during regular weather conditions, and to submerge in the water as the water current increases. In an exemplary embodiment, as the water current increases on the surface of the water, the force of the water current moves the fish cage 100 away from the position of the anchor 704. In addition, as the fish cage 100 float over the water, the wave action itself may damage the fish cage 100. However, by properly adjusting the buoyancy of the fish cage 100, the fish cage 100 may submerge in the water as the wave action moves the fish cage 100 away from the anchor 704 to a point that it stretches the upper section 702b of the mooring chain 702. For example, in an exemplary embodiment, the buoyancy of the fish cage 100 may be adjusted such that the weight of the lower section 702a of the mooring chain 702 is greater than the buoyancy of the fish cage 100. In such embodiment, as the fish cage 100 is pushed away from the anchor 704, the weight of the lower section 702a of the mooring chain 702 offsets the buoyancy of the fish cage 100 is offset by, allowing the fish cage 100 to submerge in the water as it is pushed away from the anchor 704. The buoyancy of the fish cage 100 may depend on, e.g., the volume of the fish cage 100, the density of the fish cage 100 material, the weight of the upper portion 702b of the mooring chain 702, the weight of the lower portion 702a of the mooring chain 702, etc. In an exemplary embodiment, the buoyancy of the fish cage 100 may be adapted such that, with increased wave action, as the fish cage 100 submerges in the water, it lifts a portion of the lower portion 702a of the mooring chain 702 off the seabed.

[00066] In an exemplary embodiment, the buoyancy of the frame 102 may be adjusted in several ways. In an exemplary embodiment, the frame 102 may have a tubular construction (e.g. , may be made of high density polyethylene pipes), wherein the air located inside the pipes creates the buoyancy that allows the fish cage 100 to stay afloat. In order to control the buoyancy of the fish cage 100, a ballast 720, shown, for example, in FIG. 7, may be attached to the fish cage 100. According to an exemplary embodiment, the ballast 720 may be a chain attached to the bottom of the fish cage 100, although other configurations are possible. [00067] The weight of the ballast 720 may vary depending on the size of the fish cage 100 and its desired buoyancy. In an exemplary embodiment, the weight of the ballast 720 may be based on the amount of weight needed to prevent rolling of the cage 100 and the amount of weight needed to deploy the cage 100 at near neutral buoyancy. In an exemplary embodiment, the weight of the ballast 720 may be such that it is less than, but offsets a portion of, the buoyancy of the fish cage 100. Thus, the ballast 720 helps the fish cage 100 to submerge in the water with increased wave action. Additionally, during severe storm conditions which may force the fish cage 100 deep near the seabed, the ballast 720 may help the fish cage 100 from hitting the seabed. In an exemplary embodiment, as the fish cage 100 gets closer to seabed, a portion of the ballast 720 comes in contact with seabed, causing the relative buoyancy of the fish cage 100 to increase. As the more of the ballast 720 rests on the seabed, the relative buoyancy of the fish cage 100 increases to a point that it acts against the fish cage 100 from submerging any further. Accordingly, the likelihood of the fish cage 100 itself impacting the seabed is decreased.

[00068] In addition to the ballast 720, the buoyancy of the cage 100 may be adjusted by adding air or water to the lower horizontal frame pipes 116a-116d and the vertical frame pipes 118a-118d. In an exemplary embodiment, the desired excess buoyancy of the cage 100, adjusted by the ballast 720 and the water ballast inside the frame 102, may be less than 50 kg. [00069] In a further exemplary embodiment of the invention, the buoyancy of the fish cage 100 may additionally or alternatively be adjusted by allowing water or other fluid into the frame 102. FIG. 1OA shows an exemplary design of a sloping member 118a-118d, in which water is added inside the pipe to adjust the volume of air inside the pipe. In this embodiment, the pipe 1002 may include a water plug valve 1004 arranged near the bottom of the pipe 1002, and an air plug 1006 arranged near the top of the pipe 1002.

[00070] After the fish cage 100 is placed in the water, the plug valve 1004 may be opened to allow water to flow into the pipe 1002. The plug 1006 may also be opened, allowing the air to exit the pipe 1002 as the water flows in through the plug valve 1004. Since water is heavier than air, the water in the pipe 1002 may act as a ballast, decreasing the buoyancy of the fish cage 100. By altering the level and amount of water ballast in the pipe 1002, the buoyancy of the fish cage 100 may be adjusted such that the fish cage 100 may be submerged in the water with increasing water current. In an exemplary embodiment, the water level in the vertical frame pipes may depend on the mass of the fish cage 100, as well as the amount of bio-fouling accumulated on the cage frame 102 and culture net 130, 160, 162. During the production season, the amount of water in the vertical frame pipes may be adjusted to offset changes in the total weight of the fish cage 100. In an exemplary embodiment, the excess buoyancy of the fish cage 100 may be maintained accordingly at approximately less than 50 kg. [00071] In the embodiment of the invention shown in FIG. 1OA, it may be necessary for a

SCUBA diver to operate the plug 1004 from below the water surface in order to allow the water to flow into the tube 1002. Thus, in a further exemplary embodiment of the invention shown in FIG. 1OB, an additional water plug 1010 may be arranged near the upper portion of the pipe 1002. The water plug 1010 can be connected to a water hose 1014, which can extend to near the bottom of the pipe 1002. Water may be added to the pipe 1002 through the water plug 1010. Also, the water can be taken out of the pipe 1002 by pressing pressurized air through the plug 1012. Since the water hose 1014 reaches close to the bottom of the pipe 1002, nearly all the water inside the pipe 1002 can also be removed through the plug valve 1010. In an exemplary embodiment, water may be inserted or removed from the pipe 1002 by merely lifting the fish cage 100 in or out of the water while keeping the plugs 1010, 1012 open.

[00072] According to the exemplary embodiment of the invention, the water ballast as described above may be used for manual as well as automatic submergence of the fish cage 100. For example, the water ballast may be used to allow personnel to manually submerge the fish cage 100 when a typhoon or storm is approaching. Also, the water ballast may enable the fish cage 100 to automatically submerge deeper into the water with increasing water current. Thus, as the weather conditions worsen, the fish cage 100 can automatically submerge deeper into the water, as described above, keeping farther away from the sea surface. In very severe weather conditions, the fish cage 100 may even submerge close to the seabed. Thus, the level of automatic submergence of the fish cage 100 depends on the severity of the weather, decreasing the likelihood of the fish cage 100 getting damaged by the typhoon or storm. [00073] To allow for manual submergence, according to an exemplary embodiment of the invention shown in FIG. HA, a plurality of quick release buoys 1102a- 1102d may be attached to the top portion 104 of the fish cage 100. In this embodiment, after one or more of the sloping members 118a-118d are flooded fully or partially with water to allow submergence, the quick release buoys 1102a-1102d may be attached to the top portion 104. In normal water conditions, the personnel may keep the quick release buoys 1102a- 1102d attached to top portion 104 such that the fish cage 100 floats near the water surface. Under approaching typhoon conditions, however, the quick release buoys 1102a-1102d may be removed, allowing the fish cage 100 to submerge into the water, as shown in FIG. HB. In an exemplary embodiment, sufficient water can be added to the pipes 1002 of the sloping members 118a-118d to allow the fish cage 100 to submerge 1-2 meters below the water surface. Then, as the storm intensifies and the water current increases, the fish cage 100 may sink further into the water, as depicted in FIG. HC. The fish cage 100 may submerge in the water until all or part of the ballast 720 hits the seabed. At that point, the buoyancy of the fish cage 100 could be enough to prevent the fish cage 100 itself from coming in contact with the seabed, thus preventing the fish cage 100 from getting damaged by the seabed.

[00074] Referring now to FIG. 12, an exemplary transfer mechanism for underwater fish transfer is depicted, according to an exemplary embodiment of the invention. In this embodiment, one of the netting side panels 138a-138d of the fish cage 100 may be provided with a gap 1202, which is typically sealed but may be opened during the fish transfer. A second fish cage 1210, which itself can include a gap 1212, may also be provided for the underwater fish transfer. To perform the fish transfer, a detachable tunnel 1220, formed, for example, of similar netting to the side panels 138a- 138d, can be attached to the side panel 138a- 138d over the gap 1202. The other end of the detachable tunnel 1212 can be connected to the second fish cage 1210, over the gap 1212. The gaps 1202 and 1212 are then opened, allowing the fish to swim through the tunnel 1220 and into the second fish cage 1210. The gaps 1202, 1212 may include a zipper that is sewn into the side of the net. Alternatively, the gaps 1202, 1212 may include lashing with net twine. In an exemplary embodiment, the fish may be forced out through the tunnel 1202 by detaching the bottom four corners of the fish containment net 162 and lifting the fish containment net 162, thus forcing the fish to find an exit through the tunnel 1202 opening.

Examples

EXAMPLE 1

[00075] The following example provides a more detailed analysis of various features and embodiments of the invention and also provides additional embodiments and other information. The embodiments and features disclosed in this example can be combined in various ways with each other, and with described above.

OVERVIEW

The USB OCAT cage is a submersible offshore finfish aquaculture system developed for exposed ocean locations. The cage volume is 100 cubic meters, with the cage net volume maintained by a rigid frame structure that holds the fish containment netting taut.

The OCAT cage has the shape of a truncated pyramid, with nominal dimensions of 4.5 m high, a 2.0-meter square top and a 7.0-meter square base (Figure 1 - not shown). The cage has a rigid structural frame that is an assembly of high-density extruded polyethylene pipe (HDPE) segments bolted to galvanized steel corner joints at the four top and four bottom corners. This structural framework facilitates field assembly of the non-orthogonal intersections. The bolted intersections facilitate the replacement of individual frame components should any component wear out or become damaged.

Figure 1. Photo of an OCAT cage frame prior to installation of the fish culture net and submersion of the cage to its operational depth at the offshore ASA-IM site in Lingshui Bay, China. The HDPE pipe frame is 4.5 m high, with a 2.0-meter square top that serves as a floating feed enclosure, and a 7.0-meter square base. The diagonal HDPE pipes are 5.75 m long. The HDPE pipes are bolted at each corner to galvanized steel corner weldments. A 100-m³ fish culture net is suspended inside the rigid cage frame. HDPE feed enclosure boards and a perimeter work rail surround the top of the cage frame.

CAGE FRAME HDPE Pipe

The structural frame of the OCAT cage is made from thick-walled, high-density polyethylene extruded (HDPE) pipe. The upper and lower horizontal frame members are 280 mm O. D. SDR-11 HDPE pipe with a 25.4 mm wall thickness (Figure 2). The diagonal frame members are 200 mm O. D. SDR-11 HDPE pipe with a wall thickness of 18.2 mm. The horizontal pipes intersect at 90 degrees, while the diagonal pipes are set at 52 degrees from vertical. Each HDPE frame member is sealed at each end with an HDPE plug welded to the inside of the HDPE pipe to make it air tight. Threaded valves are located at both ends of each HDPE pipe member to permit inflow and outflow of air and water into the HDPE pipes for regulating cage buoyancy (Figure 3). An internal hose attached to one of the threaded plugs allows adjustment of water and air into and out of the diagonal HDPE pipe from the water surface end of the diagonal frame members (Figures 4 and 5). Reference section "V: OCAT Engineering Drawings and Specifications" for scaled engineering drawings of the OCAT HDPE frame assembly.

Figure 2. Photo of the 280-mm SDR-11 HDPE pipe used for the horizontal lower and upper cage frame members in the original OCAT cage design. Each HDPE frame member is bolted to the galvanized steel corner units with eight galvanized or stainless steel bolt, nut and washer assemblies.

Figure 3. Photo of threaded buoyancy adjustment valves welded into the HDPE frame sections. Each HDPE pipe section is sealed at each end to make it air tight. Threaded valves allow regulation of air and water into and out of the HDPE pipes to regulate cage buoyancy. Threaded valves are protected by small square HDPE guards heat welded to the HDPE frame pipe.

Figure 4. Drawing of the threaded valve arrangement on the HDPE frame sections. Two threaded valves are included on each of the 280-mm upper and lower HDPE frame members, with one at each end on opposing sides of the HDPE pipe. Three threaded valves are included on the 200-mm diagonal HDPE frame members, two at the surface end and one at the lower end. A reinforced hose inside the 200-mm HDPE frame section is attached to the lower of the two surface end threaded valves to input and release water and air. Cage buoyancy is controlled by the amount of water and air in the diagonal frame sections. Having two threaded valves at the surface end of the diagonal frame sections allows buoyancy control from the ocean surface and eliminates the need for SCUBA support to open the lower threaded valve on the diagonal HDPE frame pipes.

Figure 5. Photo of the stainless steel and plastic threaded valve and reinforced hose to input and exhaust air and water from the diagonal HDPE frame members. The hose is installed inside the vertical HDPE frame members prior to welding the airtight plugs in the ends of the HDPE members.

Corner Units

The cage frame corner units are constructed of galvanized steel weldments. The initial 2004 corner unit design utilizes a combination of 9-inch O. D. (228.6 mm) steel pipe and 6-inch Schedule 40 steel pipe (6.33-inch O.D.;160.8 mm) that fit into the inside diameters of the 280-mm and 200-mm HDPE structural frame pipes. Each steel to HDPE connection is secured with eight bolts. There are a total of eight corner intersection weldments, four uppers and four lowers. All eight are identical except for the orientation of the diagonal tube insert sub-assembly. The fabrication of the eight steel corner units requires the assembly of two identical 90-degree intersections per corner. The 90-degree intersections are made of 9-inch steel tubing of 0.25-inch (6.35-mm) wall thickness that fit into the ends of the 280-mm diameter HDPE frame sections. Eight identical sub-assemblies for the 200-mm diameter diagonal pipes are also required and are made of 6-inch diameter Schedule 40 steel pipe that fit into the ends of the 200-mm diameter diagonal HDPE frame sections. The four top and four bottom intersections are assembled by properly aligning the smaller-diameter assemblies at 52 degrees and welding them to the larger 90- degree intersections. Reference section "V: OCAT Engineering Drawings and Specifications" for scaled engineering drawings of the OCAT corner units.

Figures 6 and 7 illustrate the upper and lower galvanized steel corner units, respectively. The 90-degree angle of the corner units is achieved in two 45-degree intersections to avoid a sharp corner hazard. The central section of the 9- inch corner intersections includes a hand hole on the outside edge to allow for the insertion and tightening of the bolts that attach the 280-mm HDPE pipe components during assembly. The central 9-inch segment also has pad- eyes drilled in flanges welded to the inside and outside edges of the central intersection segment. The pad-eyes allow for securing the fish culture net corner shackles, for attaching mooring and ballast bridles, and for other operational attachments as needed. The angled, 6-inch diameter diagonal tube segments also include a hand hole for tightening the bolts that attach the 200-mm diagonal HDPE pipe units.

Figure 6. Photo of an upper galvanized steel corner unit. Sections of 280-mm HDPE pipe bolt to the 9-inch steel tubes set at a 90-degree angle to each other. The upper end of the 200-mm diagonal frame pipe bolts to the angled 6-inch steel pipe tube that has been machined to fit into the 200-mm HDPE pipe.

Figure 7. Photo of a lower galvanized steel corner unit with the horizontal 280-mm HDPE bottom frame members bolted in place.

MOORING SYSTEM

The OCAT cage is anchored using a single point mooring (SPM). The SPM components include a concrete anchor, a mooring chain composed of three chain sizes, a surface buoy, and bridles that connect the mooring system to the cage structure as shown in Figure 8.

Figure 8. Conceptual drawing of the single point mooring (SPM) for the OCAT cage, consisting of a 5-ton concrete anchor, a three-section mooring chain, a surface buoy with near neutral buoyancy, and bridle ropes to connect the mooring system to the top and bottom corners of the OCAT cage.

Anchor

The anchor for the SPM is a 5 metric ton concrete block cast in the shape of a truncated pyramid with sloping sides angled at 45 degrees to encourage embedment should the mooring load exceed its weight-induced friction on the seabed (Figure 9). The concrete anchor block is reinforced with sufficient steel reinforcing bar to prevent breakage and to help anchor the bitter end of the 1.0-inch (25-mm) stud link mooring chain that is cast in place. Reference section "V: OCAT Engineering Drawings and Specifications" for scaled engineering drawings of the OCAT anchor.

Figure 9. Photo of the 5-metric ton OCAT anchor during deployment in Lingshui Bay, China. Mooring Chain

The mooring chain for the SPM consists of three sections: 1) a lower section of 1-inch (25-mm) stud link chain that is cast into the concrete anchor block and which normally rests on the ocean bottom, adding counterweight to aid the anchor's holding power as well as acting as a damper to the system (ref. Figure 9); 2) a mid section of galvanized 0.75-inch (19-mm) grade 3 chain which is suspended off the bottom and is connected at its bottom end to the 1-inch stud link chain with a swivel, and at its top end to a custom mooring connector with a screw pin shackle (Figures 10 and 11); and 3) an upper section of 0.5-inch (13-mm) galvanized G-40 chain that connects the custom mooring connector to a surface SPM buoy. Reference section "V: OCAT Engineering Drawings and Specifications" for specifications of the mooring chain used at the China OCAT site.

Figure 10. The mid section of galvanized 19-mm grade 3 chain is connected at its bottom end to the 25-mm stud link chain with a swivel and galvanized steel screw pin shackles. The screw pin shackles are welded shut.

Figure 11. The mid section of galvanized 19-mm grade 3 chain is connected at its top end to a custom mooring connector with a galvanized screw pin shackle. The screw pin shackle is welded shut.

Mooring Connector

The mooring connector, positioned between the 19-mm mid and 13-mm upper mooring chain segments at a water depth of 3.6 m, provides attachment for the rope bridles that link the OCAT cage to the SPM assembly. Placement of the mooring connector within the mooring chain at the proper depth ensures that the OCAT cage remains level to the ocean surface during normal operating conditions. The mooring connector is designed to minimize contact among the various SPM chains and rope bridles under constant wave action. The mooring connector is made of steel, and is a constructed weldment of four pieces: a vertical linkage, two side pad-eyes, and a spacer (Figure 12). The welded assembly is hot dip galvanized to prevent corrosion. Reference section "V: OCAT Engineering Drawings and Specifications" for further details.

Figure 12. Galvanized steel mooring connector (right) with upper and lower pad-eyes to attach the mid and upper SPM anchor chain components. Additional pad-eyes are for attachment of the one upper and two lower bridle ropes that link the OCAT cage to the SPM system. Underwater photo (below) shows attachment of the three bridle ropes that connect the OCAT cage to the mooring connector.

The SPM mooring chain is connected to a surface buoy that suspends the mid and upper sections of the mooring chain off bottom. The OCAT site in China requires a SPM buoy with a nominal buoyancy of 180 kg. The 180-kg buoyancy is slightly greater than the total dead weight of the suspended portion of the mooring chain at the 20.1 meter deep offshore site in Lingshui Bay, China. The 180-kg buoy provides near neutral buoyancy, allowing the SPM system to submerge, together with the cage, as water current increases.

The present mooring buoy for the China OCAT cages consists of a series of fifteen, 12-kg displacement globe buoys that collectively have a total buoyancy of 180 kg. The 300-mm diameter globe buoys are lashed in series to the upper, 13-mm galvanized steel chain section of the three-section mooring chain (Figure 13). A single 180-kg displacement buoy could also be used at the upper end of the mooring chain.

Figure 13. Photo of a globe buoy (above left) used for the OCAT mooring system. The OCAT cage site in China has a water depth of 20.1 m and requires fifteen of the 12-kg displacement globe buoys, attached in series to the upper section of the 0.5-inch mooring chain (above right), to provide a total buoyancy of 180 kg for the SPM system.

Bridle Ropes

The OCAT cage is linked to the SPM mooring connector by rope bridles that run from the upper and lower frame corner units of the OCAT cage (Figures 14 and 15). The rope bridals have woven eye splices in each end. The upper bridals are lashed to a single bridal extension that leads to the mooring connector. The lower bridals and the upper bridal extension are made of 1-inch (24 mm) diameter, 12-plait polyester rope with a minimum tensile strength of 27,500 lbf. The upper bridles are made of 0.75-inch (18 mm) diameter, 12-plait polyester rope with a minimum tensile strength of 24,200 lbf.

The bridles are connected to the cage corner units and the SPM mooring connector with 0.75-inch galvanized screw pin shackles. To reduce the risk of chafing, the shackles are not passed through the bridle rope eyes, but are lashed to them as shown in Figure 16. These lashings are made of polyester twine and assembled rock-hard with longitudinal strands equalling twice the strength of the rope involved. Reference section "V: OCAT Engineering Drawings and Specifications" for bridle rope specifications.

Figure 14. Plan view of the OCAT cage showing the upper and lower bridle rope assemblies. Upper bridle ropes leading from the top cage corners are lashed to a single bridle rope that leads to the SPM mooring connector. The two lower and one upper bridle ropes are lashed to screw pin shackles attached to the mooring connector.

Figure 15. Photo of the OCAT cage showing the rope bridles attached to the bottom and top steel corner units of the OCAT cage. The bridles connect the cage to the SPM.

Figure 16. Photo showing a bridle rope lashed to a screw pin shackle that is attached to the padeye of the steel corner unit of the OCAT cage frame. The screw pin of the shackle is wired shut to prevent loss of the shackle and bridle attachment.

FEED ENCLOSURE

The 2.0-meter square HDPE top frame of the OCAT cage forms the structural base for a feed enclosure designed to contain extruded, floating feed pellets. The feed containment area is enhanced by protective splashboards made of 18-mm HDPE plate (white plates pictured in Figure 17). Each splashboard section is bolted to three 110-mm HDPE structural tubes welded vertically to the top surface of the upper 280-mm HDPE frame (Figure 18). The HDPE structural tubes also serve as the base for 110-mm diameter x 80-cm long HDPE posts that support a top guard rail that encompasses the feed enclosure (Figures 19 and 20). The bottoms of the guard rail posts are machined down so that they insert into and are bolted to the vertical structural tubes. The open corner areas between the four splashboards are closed with HDPE plate material bolted to the splashboards (Figures 21-23). The corner HDPE plate material for the OCAT cages in China was cut from a 200-liter plastic barrel.

Figure 17. Photo of the top feed enclosure of the OCAT cage. The white splashboards form a perimeter feed enclosure to help contain floating feed pellets and are made of 18-mm HDPE plate bolted to HDPE structural tubes that are welded to the top surface of the 280-mm HDPE frame members. The HDPE splashboards were originally attached to the inside of the structural tubes, but were later moved to the outside of the structural tubes to eliminate chaffing to the fish culture net.

Figure 18. Three 110-mm HDPE structural tubes are welded vertically to each of the four 280-mm HDPE upper frame members. The structural tubes form the base for the posts of the guard rail and for bolting the HDPE feed enclosure splashboards.

Figure 19. A guard rail constructed from 110-mm HDPE pipe surrounds the top of the OCAT cage. The guard rail is attached by inserting the three rail posts of each rail section into the structural tubes welded to the 280-mm HDPE frame members and bolting the rail posts in place.

Figure 20. Photo of a section of the OCAT guard rail. The 80-cm high rail is constructed of 110-mm HDPE pipe. The bottom section of each HDPE rail post is machined so that the post can be inserted into and bolted to the 110- mm structural tubes welded to the top of the 280-mm HDPE cage frame.

Figure 21. Open voids between the HDPE feed enclosure boards are closed to prevent feed loss. Upper and lower splashboards (white) were initially installed to form the feed enclosure. The lower splashboards were later removed to reduce net chaffing.

Figure 22. The open voids between the HDPE feed enclosure boards on the China OCAT cages are closed with plastic material cut from a locally procured plastic barrel.

Figure 23. Flexible HDPE plate material cut from a plastic barrel is bolted to the white HDPE splashboards to close the corner voids of the top feed enclosure.

FISH CONTAINMENT NET - Main Containment Net

The main fish culture containment net is an assembly of five netting panels and a series of framing lines lashed together to match the truncated pyramid shape of the cage frame (Figure 24). The four side panels are identical trapezoids, while the bottom panel is square (Figure 25). Each panel is edged with a perimeter line to facilitate individual replacement. All panels are then seized to the framing lines. An upper addition to the main containment net extends the net to the top of the feed enclosure rail system (Figure 24).

Figure 24. Schematic drawing of the main fish containment net showing the net dimensions and mesh counts for the side panels and the top extension. The containment net is made of 3.0-cm square mesh polyethylene netting.

Figure 25. Schematic drawing of the bottom panel of the main fish containment net showing the net dimensions, mesh counts and perimeter and framing line twine sizes. The main containment net and upper net extensions are made of 3.0-cm square mesh (6-cm stretch mesh) polyethylene netting (Figure 26). The netting is made from 2.5-mm diameter twine, formed from two 30-string strands twisted together. The netting is manufactured by Asahi Kasei Fiber Company. For the China OCAT cages, the netting was purchased and the cage nets sewn by a local net maker in Xincun, China, near the OCAT project offshore cage site.

Figure 26. Photo of the 3.0-cm square mesh polyethylene netting material used for the main containment net and upper net extension.

The four side and one bottom main containment net panels are lashed together with 2.5-mm polyethylene twine. The perimeter two meshes of each panel are drawn together and lashed through each 3.0-cm mesh square to bind the panels together (Figure 27). An inner perimeter line is then added to each panel by threading 10-mm diameter nylon rope through the perimeter meshes of each panel (Figure 28). The inner perimeter lines of the four diagonal side seams of the containment net are then lashed to 13 -mm outer nylon framing lines (Figure 29). The framing lines are 10 cm longer than the mating panel edge and have 10-cm interior length eyelets that extend partially beyond the netting (Figure 30). The framing lines strengthen and tension the net and secure the net to the cage frame. The four top corner framing line eyelets are shackled directly to the frame's top corner unit pad-eyes with galvanized screw-pin shackles. The four bottom corner framing line eyelets have screw-pin shackles that are lashed to a second screw-pin shackle that is attached to the steel corner unit pad-eye of the frame.

Figure 27. Photo showing the perimeter two meshes of each panel drawn together and lashed through each 3.0-cm mesh square. The four side panels and one bottom panel are lashed together in this manner to form the main containment net.

Figure 28. A 10-mm diameter polyethylene line is passed through the perimeter mesh of each net panel seam to form an inner perimeter line.

Figure 29 The 10-mm diameter inner perimeter lines are lashed to 13 -mm framing lines The framing lines allow attachment of the containment net to the cage frame

Figure 30 Photo of the framing lines Each of the four diagonall3-mm framing lines has a 10-cm long eyelet at each end for attaching the containment net to the cage frame The framing line eyelets at the top of the containment net are attached directly to the upper frame corner unit pad-eyes with galvanized screw-pin shackles The framing line eyelets at the bottom of the containment net have individual galvanized screw-pin shackles that are lashed to additional galvanized steel screw pin shackles attached to the lower corner unit pad-eyes (ref Figure 16 for lashing example) The shackle to shackle lashings permit tensioning of the net

Top Net

The top of the cage is covered with a separate net panel to provide a completely enclosed net system to prevent fish escapement This net is made of 1 6-cm square mesh (3 0-cm stretch) polyethylene netting, lashed around the perimeter to a 1 2-cm, 3-strand nylon rope (Figure 31) The perimeter rope of the top net is lashed to the perimeter rope of the top extension of the main containment net, and then lashed to the top rails to keep the net taut (Figure 32) The 1 6-cm square mesh netting allows feed to be poured through the top net to the water surface below

Figure 31 The 1 6-cm square mesh top net is lashed at each mesh to a 1 2-cm, 3-strand perimeter line

Figure 32. The perimeter lines of the top net and the top extension of the main containment net are lashed together to form an escape proof containment net. The perimeter lines are then attached to the cage frame with 8-mm, 3- strand nylon ropes that lead from the four corners of the top net to the cage frame corner pad-eyes. The top net perimeter line is also lashed to the top rails to keep the top net taut.

Feed Enclosure Net

A feed enclosure net is attached inside the main containment net at the top of the cage to contain floating feed. The feed enclosure net is made from 3-mm square mesh polyethylene netting and has the same shape as the main containment net and top extension (Figure 33). The feed enclosure net extends above and below the water line as indicated in Figure 33.

Figure 33. The feed enclosure net extends above and below water to prevent floating feed pellets from being lost as water moves through the cage, or during vigorous feeding activity.

The feed enclosure net is lashed to a 2.5-mm nylon inner perimeter line and a 5-mm nylon outer perimeter line with a stitch line (Figure 34). The top and side, and the bottom and side, perimeter lines are lashed together, respectively, at the corners and nylon rope extensions are attached to tie the top and bottom corners of the feed enclosure net to the main containment net to keep the feed enclosure net taut (Figure 35).

Figure 34. The feed enclosure net is lashed to inner and outer perimeter lines to tension the feed enclosure net and secure it to the main containment net.

Figure 35. Nylon rope extensions are attached at the top and bottom corners of the feed enclosure net where the perimeter lines join.

Fish Transfer Net

A tubular shaped tunnel net is used to transfer fish from the OCAT cage to an adjacent live car during net changing and fish harvest activities. The tunnel net is made of 3-cm square polyethylene netting and is sewn into a 95-cm diameter and 5-m long tube (Figure 36).

A 95-cm diameter stainless steel hoop is sewn to each end of the tunnel net to facilitate attachment of the tunnel net to the OCAT cage on one end and a live car fish holding cage on the other end. Corresponding 95-cm diameter stainless steel hoops are lashed to the bottom portion of the down-current side of the OCAT fish containment net and near the bottom edge of one side of the live car holding cage net (Figure 37). The hoop openings are either closed with a zipper or lashing.

Figure 36. Photo of an OCAT fish transfer tunnel net. The tunnel net is a 95-cm diameter by 5-m long net tube, with a 95-cm diameter stainless steel hoop attached at each end of the netting tube to facilitate attachment to the OCAT cage on one end and to a fish holding cage on the other end.

Figure 37. Underwater photo showing the 95-cm diameter stainless steel hoop sewn into the lower portion of the side panel of the OCAT fish containment net. The 95-cm diameter circular opening inside the hoop can be closed with a zipper or lashing.

CAGE BALLAST

Ballast is provided by a pendant chain suspended below the center of the OCAT cage (Figure 8). The China OCAT cage ballast is made from a section of 28-mm steel stud-link chain weighing 96 kg (Figure 38). The ballast chain is suspended beneath the center of the OCAT cage from equidistant rope bridles that lead from the four corners of the cage and are connected as a group to the ballast chain (Figure 39). The low center of gravity of the ballast chain aids in the stability of the cage. It also assists in stopping the descent of the cage in extreme current conditions so that the cage does not touch the ocean floor and sustain potential damage (Figure 40). The size and length of the ballast chain are non-critical and its weight can be adjusted to assist in submerging the cage to its optimal operational position.

Each of the four ballast bridles are 7.0 m long, and are made of 18-mm nylon rope with eye splices in each end. The construction of these lines can be braided, plaited, or three-strand. Like the mooring bridles, the ballast bridles are lashed to galvanized steel shackles at the point of attachment to steel pad-eyes in the four lower corner units of the cage frame.

Figure 38. Photo of a section of the 28-mm stud-link chain used as ballast in the China OCAT cages. Each link weighs approximately 4.3 kg.

Figure 39. The four, 7.0-m long ballast bridles are connected with galvanized steel screw pin shackles so that they hang from the four corners of the cage frame to a center point beneath the cage.

Figure 40. Diagram of the submersed OCAT cage descent to the ocean floor. The ballast is designed to stop the descent of the cage in extreme current conditions so that the cage does not touch the ocean floor and sustain potential damage.

CAGE OPERATIONS AND MAINTENANCE CAGE BUOYANCY AND STABILITY

The OCAT cage is designed so that floatation buoyancy is provided by the HDPE cage frame. When all HDPE cage frame members are filled with air, the cage frame floats on the water surface (Figure 41). This simplifies transfer of the cage from the onshore assembly site to the offshore operational site. The China OCAT cages were assembled on a local beach in Xincun, Hainan, China, and towed on the ocean surface by a local fishing boat to the installation site 4 km offshore (Figure 42).

Once on site at the offshore operational location, the lower 7.0-m x 280-mm HDPE cage frame members are flooded with water by opening the buoyancy adjustment valves at each end of each HDPE pipe member (ref. Figure 3). Flooding of the lower HDPE cage frame ring lowers the cage in the water to near its operational depth (Figure 43). Operational buoyancy is then attained by partially flooding the four diagonal HDPE cage frame members with water until the OCAT cage is submerged with only the top half of the 2.0-m x 2.0-m upper HDPE frame ring exposed (Figure 44). Cage buoyancy can be periodically adjusted to respond to the added weight from frame and net bio-fouling, or loss of air from the HDPE frame members, by adjusting the amount of air and water in the four diagonal HDPE frame members. Optimal operational buoyancy provides <250-kg of excess buoyancy, so that an increase in water current velocity to approximately 50 cm/sec will begin to submerge the cage.

Cage stability within the water column is provided by the combination of the ballast suspended beneath the OCAT cage and the flooded lower HDPE frame ring. This low center of gravity prevents the cage from rolling or tipping over.

Figure 41. Photo of a newly assembled OCAT cage frame with the HDPE frame members filled with air. When filled with air, the cage frame floats on the water surface.

Figure 42. Photo of an OCAT cage being towed to its offshore site by a local Chinese fishing boat.

Figure 43. Photo of an OCAT cage being submerged at the offshore installation site by flooding the lower 7.0-m x 7.0-m HDPE frame members with water.

Figure 44. Photo of a fully deployed OCAT cage with approximately one-half of the upper HDPE frame ring exposed above the water surface (photo taken before installation of the cage rail system).

CAGE SUBMERSION

The OCAT cage is designed to operate with the majority of the cage submerged below the water surface, with only the upper half of the upper 280-mm HDPE frame ring exposed above the water surface, together with the upper feed enclose boards and rail system (reference Figure 32). This standard operational position permits daily feeding of the fish at the surface by dispensing feed within the top feed enclosure of the cage. At the OCAT cage site in Lingshui Bay, China, normal daily sea conditions have an average wave height of approximately 1.0-m and a water current velocity of 0.15-0.20 m/sec. The OCAT cages deployed at this site remain continuously at the water surface under these conditions.

Submersion of the OCAT cage can be attained either automatically or manually. Automatic submersion is designed to occur when storm generated increases in water current velocity exert sufficient pressure on the up-current side of the cage to force downward movement of the cage. Numerical modeling tests conducted at the University of New Hampshire (UNH) indicate that nominal submergence of the OCAT cage will begin at a water current velocity of 0.25 m/sec, and that complete submergence of the cage top rail (cage submergence to 0.77-m depth) will occur at a water current velocity of approximately 0.75 m/sec (Table 1). At a water current velocity of 1.5 m/sec, the cage is predicted to submerge approximately 16 m below the ocean surface. Resurfacing of the OCAT cage occurs when water current velocity returns to normal. The full report of the UNH numerical modeling study includes impact assessments of varying sea and current conditions on the OCAT cage system and is summarized in "Section VI: Numerical Modeling".

Table 1. Depth response of the OCAT cage system in 24 meter deep water with applied increasing current regimes (Celikkol et at., 2007a).

At-sea verification of the submergence depths shown in Table 1 has not been conducted at the offshore OCAT site in China. The four China OCAT cages, however, weathered a series of typhoon-strength storms during the 2005, 2006 and 2007 feeding trials. The most severe of these storms, typhoon Damrey in September 2005, had peak winds of 125 mph (200 kph) and sustained winds at the OCAT cage site of 80 mph (130 kph). These conditions are believed to have generated surface water current conditions of up to 2.0 m/sec. Two OCAT cages were on site at the time of typhoon Damrey. One of the two OCAT cages had no water in its diagonal frame members, while the diagonal frame members of the second OCAT cage were partially flooded to reduce cage buoyancy. The cage with only air in the diagonal frame members remained at the ocean surface throughout the storm and sustained minor damage. Damage consisted of loosening of the upper frame bolt assemblies and breakage of one of the HDPE feed enclosure boards. Pompano being cultured in this cage also exhibited external physical abrasions that resulted in post-storm fish losses. The OCAT cage that had the diagonal frame members partially flooded submerged and sustained no visible storm damage to the cage or fish. At-sea testing of OCAT cage automatic submergence is scheduled in the Gulf of Maine by UNH in 2008, using a prototype OCAT cage constructed by UNH with U.S. soybean industry funding (ref. section IX: OCAT Design Evaluation and Modified Cage Construction).

Manual submergence of the OCAT cage is accomplished by adjusting the wateπair ratio in the diagonal HDPE members of the OCAT cage frame. Adjusting the wateπair ratio allows positioning of the OCAT cage at any water depth from the surface to an at rest position on the ocean floor. Manual submergence has been verified through at- sea testing at the OCAT cage site in China. An air adjustment manifold connected to an on-board air compressor is used in China to adjust the wateπair ratio in the diagonal cage frame members. Air lines lead from the manifold to the threaded valves on each of the four diagonal frame members. Valves on each air line at the air manifold allow air to be inputted or exhausted from the diagonal cage frame members.

The OCAT cage can be operated from a fully submerged position as well as at the ocean surface. Performance testing of an OCAT cage under continuous submergence conditions is scheduled in 2008 at the UNH offshore cage grid system in the Gulf of Maine, with funding from the U.S. soybean industry. The UNH offshore cage site was designed to conduct engineering assessment and fish culture studies in submerged SeaStation cages, but the grid system has the capability to insert additional cage designs for testing purposes. NET CHANGING AND CLEANING

Net changing and cleaning is required at the beginning and end of production cycles and periodically during the production season to clean nets clogged with bio-fouling organisms. Net changes with the OCAT cage are accomplished by first transferring the fish out of the OCAT cage, via a tunnel net, to an adjacent fish holding cage. Transferring of the fish by use of a tunnel net eliminates handling of the fish and minimizes fish stress. To transfer fish, the stainless steel hoops at each end of the tunnel net are lashed by divers to corresponding hoops sewn into the OCAT cage net and the holding cage net. The netting inside the stainless steel hoops is then opened to allow fish passage from the OCAT cage through the tunnel net to the holding cage (Figure 45). The hoop openings on the China OCAT cages are lashed closed with net twine, but zippered openings can also be installed to facilitate quicker opening and closing of the hoop netting. Complete removal of fish from the OCAT cage is accomplished by releasing the bottom four corners of the OCAT cage net from their attachment points at the lower four frame corner units, followed by lifting of the OCAT cage net to encourage the fish to swim out through the tunnel net opening. This process is reversed to move fish from the holding cage back into the OCAT cage. The process requires approximately 20-30 minutes once the tunnel net is in place and open at both ends.

Bio-fouled OCAT cage nets are taken to shore for cleaning with a high pressure hose. A minimum of one spare OCAT cage net is required to allow for changing and cleaning of nets during the production season.

Figure 45. Photo of pompano moving through a tunnel net from the OCAT cage to an adjacent holding cage during harvest.

The OCAT cages in China use a 95-cm diameter tunnel net to transfer fish into and out of the OCAT cages. The tunnel net diameter was based on the availability and size of locally available stainless steel hoops in China. Due to the density of fish moving through the tunnel net at one time, a larger diameter (1.5-m to 2.0-m) tunnel net is recommended to reduce fish stress inside the tunnel net and to speed fish transfers. Improvement in the procedure for attaching the stainless steel hoops at the ends of the tunnel net to the OCAT and holding cages is also recommended to reduce diver requirements for this procedure. A quick connect coupling would facilitate more rapid attachment and release of the attachment hoops by divers.

BIO-FOULING Cage Frame

The frames and nets of the OCAT cages initially deployed in China were not treated to reduce bio-fouling. Bio- fouling subsequently became a significant problem, particularly on the OCAT cage frames and buoys, and with nylon netting (Figures 46 and 47). To reduce bio-fouling and frame cleaning requirements, the frames of the China OCAT cages were painted in 2007 with a high quality marine boat anti bio-fouling paint (Figure 48). After painting the OCAT cage frames with anti bio-fouling paint, there was minimal bio-fouling of the cage frames during their offshore deployment from July to November 2007 (Figure 49). Painting of OCAT cage frames with a high quality boat anti bio-fouling paint is recommended prior to initial deployment, and at periodic intervals to maintain a high degree of bio-fouling protection. Painting of the mooring system buoy(s) with anti bio-fouling paint is also recommended.

Figure 46. Bio-fouling by a variety of marine organisms was a significant problem on the OCAT cage frames (left) and mooring buoys (right) prior to being painted with a high quality marine boat anti bio-fouling paint.

Figure 47. Bio-fouling was hand scraped from the OCAT cage frames annually prior to painting the frames with anti bio-fouling paint.

Figure 48. The OCAT cage frames were cleaned and painted with marine boat anti bio-fouling paint prior to the 2007 production season.

Figure 49. OCAT cages deployed from July through November 2007 experienced minimal bio-fouling following painting with marine boat anti bio-fouling paint.

Culture Net

The rapidity and degree of net bio-fouling varied during the 2004 through 2007 feeding trial deployments of the OCAT cages in China, depending on the type of netting and netting mesh size used. Smaller mesh netting bio- fouled quicker and to a greater degree than larger mesh netting. Knotted, uncoated nylon netting, specified by the original OCAT design engineer, stretched and sagged and became heavy and difficult to handle due to the high degree of bio-fouling. Changing nylon nets was difficult due to their weight and the extent of bio-fouling. Stretching also made the nylon netting difficult to lash to the net framing lines during fabrication.

Polyethylene netting was found to be superior to nylon netting in all aspects. The 3.0-cm square mesh polyethylene netting currently used on the China OCAT cages experiences less bio-fouling than the earlier nylon netting. No net changes due to bio-fouling were required with the 3.0-cm mesh polyethylene netting during the July-November 2007 deployment of the OCAT cages in China. The required frequency of net cleaning with the 3.0-cm mesh polyethylene netting appears to be once every 4-5 months under the temperate climate conditions prevalent at the China OCAT site. Net cleaning frequency will vary depending on site conditions.

CAGE MAINTENANCE HDPE Frame Assemblies

Loosening of the bolt assemblies that connect the HDPE frame members to the galvanized steel corner units of the OCAT cages was found to occur in response to the constant wave and current racking conditions characteristic of offshore ocean locations. The bolt assemblies that connect the 200-mm diagonal HDPE frame members to the upper corner units, in particular, are subject to loosening. The significant wave action at the ocean surface, intensified by the resistance against the feed enclosure, creates deflections and displacement of much greater magnitude on the top frame compared to the bottom. Bolt assembly loosening can result in wallowing and potential failure of the HDPE bolt holes. Frequent monitoring of the HDPE corner unit bolt assemblies is required to prevent bolt assembly loosening, HDPE hole wallowing, and potential frame failure due to bolt assembly loss. One incident of frame failure occurred in March 2008 after an OCAT cage frame, without a net, was left unattended at the China offshore site for approximately four months. The cage had been in operation for four years, and had experienced repeated loosening of the upper rim frame bolts. During the winter months in early 2008, loosening of the frame bolts and wallowing of the bolt holes resulted in the upper cage frame rim disconnecting from the diagonal frame members (Figures 50 and 51). New HDPE extensions had to be welded to the four 200-mm frame members and welding repairs made to the four upper galvanized steel corner units. Monthly checking and tightening of the 200-mm upper diagonal HDPE bolt assembly connections is considered essential to maintaining frame integrity. Loosening of the HDPE corner unit bolt assemblies for the upper 280-mm HDPE frame ring was found to occur infrequently in comparison to the 200-mm diagonal HDPE connections. No loosening of the bolt assemblies in the lower frame corners has been noted.

The quality of locally available galvanized and stainless steel bolt assembly hardware in China was found to be poor. Locally manufactured bolt assemblies, screw pin shackles and other cage hardware rusted and decayed much quicker than the same hardware manufactured in the U.S. Replacement of all locally manufactured galvanized and stainless steel hardware was found to be required annually. The OCAT cages in China are typically towed to shore at the end of each production season for replacement of in-country manufactured bolt assemblies and shackles. This refurbishment requires approximately one day per cage. Each bolt assembly and shackle is replaced in sequence with new hardware so that disassembly of the cage frame is not required. Bridle Ropes

Nylon bridle ropes are subject to bio-fouling that can weaken the rope fibers. Barnacle attachment is particularly damaging. Bridle ropes should be inspected annually for bio-fouling damage and replaced as required. Replacement of mooring bridle ropes was found to be necessary approximately every two years with the China OCAT cages. The ballast bridle ropes on the China OCAT cages have sustained less bio-fouling than the mooring bridle ropes, and have not had to be replaced during four years of operation in China.

Figure 50. The top frame rim of a four-year old OCAT cage separated from the diagonal HDPE frame members in March 2008 due to bolt loosening and hole wallowing after the cage was left unattended offshore for four months. The cage frame was towed to shore and new HDPE extensions welded to the upper ends of the four diagonal HDPE frame members. Fish were not being cultured in the cage at the time of the rim separation.

Figure 51. The upper frame rim that separated from the OCAT cage in March 2008 remained intact and was towed to shore for repairs and reattachment. Wallowing of the holes in the galvanized steel corner units required welding repairs.

OCAT ENGINEERING DRAWINGS AND SPECIFICATIONS CAGE FRAME

The 100-m³ OCAT cage net volume is maintained by a rigid cage frame made of HDPE pipe connected at the corners to steel weldments. The rigid frame consists of twelve sections of HDPE pipe: four 2.0-m upper sections, four 5.75-m diagonal sections, and four 7.0-m lower sections.

The upper and lower frame sections are made from 280 mm O. D. SDR-I l HDPE pipe with a 25.4 mm wall thickness. The lower HDPE sections have a finished pipe length of 7.0 m (Figure 1). HDPE plugs are welded into one end of 50 cm HDPE pipe extensions. The extensions are then butt welded onto each end of a 6.0 m section of HDPE pipe (Figure 1). To allow for pressure testing, flooding, and deballasting, the air tight center portion is fitted with two threaded valve inserts located close to the end plugs and on opposite sides off the pipe.

Figure 1. X-section diagram of the original lower HDPE frame sections. The frame sections are made of 280 mm O. D. HDPE pipe with a wall thickness of 25.4 mm. Four 7.0 m long frame sections are required for the square, 7 m x 7 m lower horizontal portion of the OCAT cage frame.

The upper HDPE frame sections have a finished pipe length of 2.0 m (Figure 2). HDPE plugs are welded 15 cm from the pipe end of 30-cm HDPE pipe extensions. The 30-cm pipe extensions are then butt welded to each end of a 1.4 m section of HDPE pipe (Figure 2). To allow for pressure testing, flooding, and deballasting, the air tight center portion is fitted with two threaded valve inserts located close to the plugs and on opposite sides off the pipe.

Figure 2. X-section diagram of the upper HDPE frame sections. The frame sections are made of 280 mm O. D. HDPE pipe with a wall thickness of 25.4 mm. Four 2.0 m long frame sections are required for the upper horizontal portion of the OCAT cage frame.

The side diagonal pipes| are shown in Figure 3. The diagonal frame sections are made from 200 mm O. D. SDR-11 HDPE pipe with a wall thickness of 18.2 mm. The diagonal HDPE members have a finished pipe length of 5.75 m. The finished pipe is made by welding HDPE plugs at the midpoint of 30-cm HDPE pipe extensions. The 30-cm pipe extensions containing the plugs are then butt welded onto each end a 5.15 m section of HDPE pipe (Figure 3). To allow for pressure testing, flooding, and deballasting, the air tight center portion is fitted with three threaded valve inserts (Figure 4). Two of the threaded inserts are located close to the pipe plug at the surface end of the pipe, while the third threaded insert is located at the opposite end and on the opposite side of the pipe. A reinforced hose is attached to one of the surface end threaded inserts (Reference Section III, Figure 5). The two threaded inserts at the upper end of the diagonal HDPE pipes allow adjustment of the cage buoyancy, through adjustment of the amount of air or water in the diagonal frame member, from the ocean surface without having to dive to the bottom of the cage to open the lower threaded insert (Figure 4).

Figure 3. X-section diagram of the diagonal HDPE frame sections. The frame sections are made of 200 mm O. D. HDPE pipe with a wall thickness of 18.2 mm. Four 5.75 m long frame sections are required for the angled, diagi ie OCAT cage frame.

Figure 4. X-section diagram of the diagonal HDPE frame pipes. Three threaded valves are heat welded into the 200-mm diagonal HDPE frame sections, two at the upper, surface end of the diagonal HDPE pipe, and one at the lower end. An HDPE pipe/hose inside the 200-mm HDPE frame members is attached to one of the surface end threaded valves to input and release water and air from the diagonal pipe section. Cage buoyancy is controlled by the amount of water and air in the diagonal frame sections. Having two threaded valves at the surface end of the diagonal frame sections allows buoyancy control from the ocean surface without diving to the bottom of the cage.

FRAME CORNER UNITS

The OCAT cage has a rigid structural frame that is an assembly of high-density extruded polyethylene pipe segments bolted to galvanized steel corner joints at the four top and four bottom corners. This method of assembly facilitates field assembly of the non-orthogonal intersections while providing the structural integrity of a welded polyethylene pipe structure. Figure C-I shows an engineering drawing of the 2004 galvanized steel corner unit design (Goudey, 2004).

Figure C-I Engineering diagram of the original design for the lower galvanized steel corner unit for the OCAT cage Four lower and four upper corner units are required The angle of the diagonal 6-inch tubular pipes is reversed on the four upper corner units

The following engineering drawings provide component details for the construction of the original galvanized steel corner units of the OCAT cage, beginning with an assembly drawing of the 90-degree fabricated corner unit (Figure C-2) (Goudey, 2004)

Table C-I lists the components required for each corner. The assembly drawing for the side pipe inserts is shown in Figure C-8 (Goduey, 2004). The component details for the side pipe inserts are shown in Figures C -9 through C-12 (Goudey, 2004).

Figure C-3. Steel two-hole pad-eye. Three are required for each of the eight OCAT cage frame corners.

Figure C-4. Details of the 45° transition tube section of the steel corner weldment. One transition tube is required per corner.

Figure C-5. Details of the 22.5° angled insert tube for the OCAT cage steel corner unit. Two angled insert tubes are required per cage corner unit.

Figure C-6. Steel disk that will form the outer butt end of each 9-inch steel tube that will insert into the 280-mm HDPE frame pipes. Two steel disks are required per cage corner unit.

Figure C-I . Steel flange that is welded around the 9-inch, 22.5° angled insert tube to form a butt face for the HDPE frame pipe that the 9-inch tube will fit into. Two steel flanges are required per cage corner unit.

Table C-I. Materials list for one OCAT galvanized steel corner weldment. Eight corner weldments are required per cage.

Figure C-8. Assembly drawing of the side 6-inch steel pipe insert for the OCAT cage corner unit.

Figure C -9. Corner pipe mounting plate. One required per corner.

Figure C-IO. Steel disk for butt end Figure C-I l . Steel flange for 6-inch of 6-inch angled pipe insert. One angled pipe insert. One required per required per corner. corner.

Figure C- 12. Steel 6-inch side tube insert. One required per corner.

The completed side assemblies are welded onto the 90° corner assembly oriented at 45° to the two main frame pipes. The entire corner assembly weighs 46 kg. When hot-dip galvanized, the weight of each unit increases to approximately 47.6 kg.

SINGLE POINT MOORING

The OCAT cage is anchored using a single point mooring (SPM). The SPM components include a 5-ton concrete anchor, a mooring chain composed of three chain sizes, a surface buoy, and bridle ropes that connect the mooring system to the top and bottom corners of the OCAT cage as shown in Figure C-13.

Figure C-13. Conceptual drawing of the single point mooring (SPM) for the OCAT cage.

Anchor

The OCAT anchor is a reinforced concrete mooring block in the shape of a square truncated pyramid. The overall dimensions of the mooring block are 2.44 m square by 0.61 m high. The four sides of the block slope at 45 degrees and the top of the block truncates at a 1.22 m square. This geometry yields a volume of 2.12 cubic meters and a dry weight of 5.0 metric tons. Construction details are shown in Figures C-14 and C15.

Figure C-14. A plywood form is used to cast the concrete mooring block. The stud-link mooring chain and re-bar are supported from above and kept central to the block.

Figure C-15. Engineering details of the 5.0-mt steel reinforced concrete mooring block for the SPM of the OCAT cage (Goudey, 2004).

The 1-inch stud link bottom chain is cast into the concrete mooring block at the time of its manufacture and protrudes from the top center of the block. The shape is designed to have similar holding powers in all directions of pull. The lower edge of the block should be cast sharp to encourage penetration of the block into the ocean bottom were it to move under extreme load.

The concrete casting is reinforced with steel reinforcing bar (re-bar) to keep the block intact in case of rough handling and to provide a means of securing the mooring chain end. The armature is composed of two layers of re- bar, each supported by and passing through the links of the embedded mooring chain. The lower level of re-bar is placed 10 cm from the bottom of the block. The upper level is approximately 20 cm from the top of the block. Each layer is made up of two 19-mm re-bars that pass through the chain links and are oriented to the diagonals of the block. In addition, four 10-mm re-bars are arranged in a square at their outer ends, aligned parallel to the edges of the block. The intersections of all re-bar should be secured with wire. During the pour, all re-bar should be kept at least 5 cm from the edge of the block. The re-bar materials required for one mooring block are listed in Table C-2. Table C-2. Steel reinforcing bar requirements for the OCAT 5-mt concrete mooring block.

Mooring Connector

The mooring connector provides attachment for the rope bridles that link the OCAT cage to the SPM assembly. The mooring connector is made of steel, and is a constructed weldment of four pieces: a vertical linkage, two side pad-eyes, and a spacer (Figure C-16). The welded assembly is hot dip galvanized to prevent corrosion.

Figure C- 16. Engineering drawing of the mooring connector assembly and component parts (Goudey, 2004). Mooring Chain

The mooring chain for the SPM consists of three sections: 1) a lower section of 1-inch stud link chain which normally rests on the ocean bottom and that is cast at one end into the concrete mooring block; 2) a mid section of galvanized 0.75-inch grade 3 chain which is suspended off the bottom, and is connected at its bottom end to the 1- inch stud link chain with a swivel, and at its top end to a custom mooring connector; and 3) an upper section of 0.5- inch galvanized G-40 chain that connects the custom mooring connector to the surface SPM buoy (reference Figures 10 and 11, page 13). Table C-3 shows the materials list for the mooring chain and connectors. Table C-3. Materials list for the mooring chain and connectors required for a site with a 20-meter water depth. The length of the mooring chain sections will vary with water depth.

Among the materials listed in Table C-3, the 1-inch galvanized screw pin shackle connects the 1-inch galvanized jaw/jaw swivel to the 1-inch bottom stud link chain. One of the 0.75-inch galvanized screw pin shackles connects the 1 -inch jaw/jaw swivel to the 0.75-inch galvanized off-bottom chain. The other eight galvanized 0.75-inch screw pin shackles connect the 0.75-inch galvanized off-bottom chain to the custom mooring connector, the two lower bridle ropes and the one upper bridle rope extension to the custom mooring connector, and the two upper and two lower bridle ropes to the top and bottom corners of the OCAT cage. The two galvanized 0.5-inch screw pin shackles connect the 0.5-inch buoy chain to the custom mooring connector and the surface buoy.

Mooring Bridles

The OCAT cage is linked to the SPM mooring connector by rope bridles that run from the upper and lower frame corner units of the OCAT cage. The rope bridals have woven eye splices in each end. The upper bridals are lashed to a single bridal extension that leads to the mooring connector. Bridle dimensions are shown in Table C-4 and are the dimensions from the inside of the eye splices at each end.

Table C-4. Quantity and dimensional specifications for the mooring bridle ropes that connect the OCAT cage to the SPM.

The lower bridals and the upper bridal extension are made of 1-inch (24 mm) diameter, 12-plait polyester rope with a maximum tensile strength of 27,500 lbf (Table C-5)¹. The upper bridles are made of 0.75-inch (18 mm) diameter, 12-plait polyester rope with a maximum tensile strength of 24,200 lbf. Safe working loads for the mooring bridles were determined using the standard accepted practice of dividing the maximum tensile strength by a safety factor

¹ Maximum tensile strength information was obtained from the manufacturer' s web site from 5 to 7 (Schmittou, 2005). This safety factor allows for weathering or incidental damage to the rope. The more conservative of these factors yields safe working loads for the OCAT bridle ropes of 3,457 lbf and 3,929 lbf (Table C-5).

The maximum anticipated loading on the mooring bridle ropes was determined by the UNH AquaFE Model (Celikkol et al., 2007b). The AquaFE model is capable of predicting time dependent change in position of specified nodes and corresponding internal stress (tension) within any of the component members. Worse case typhoon conditions were considered in this simulation, consisting of 7-m waves at period of 9.6 seconds accompanied by a current velocity of 2 m/s. Results are summarized in Table C-5. Details are reported in "Section VI. Numerical Modeling". Based on the modeling results, the 1.0-inch and 0.75-inch mooring bridle ropes were considered to have sufficient tensile strength for known conditions at the China OCAT trial site off southern Hainan Island. - s (Schmittou, 2005).

Cage Assembly Hardware

Assembly of the cage and the mooring systems requires the hardware components listed in Table 6. Substitution of Chinese manufactured stainless steel and galvanized hardware of the same size specifications presented no difficulties, other than wearing as reported in "Section IV. Cage Operations and Maintenance", during feeding trials conducted from 2005 through 2007.

Table 6. Hardware requirements for assembly of the OCAT cage frame and mooring system. All specifications are in English units and include the U.S. McMaster-Carr part number.

Details of the 4-piece bolt (bolt, flat washer, lock washer and nut) assembly used to attach the HDPE frame members to the steel corner units is shown in Figure C-17. Eight bolt assemblies are required for each HDPE to steel corner unit connection. There are a total of 24 HDPE to steel corner unit connections requiring 192 bolt assemblies.

Figure C-17. X-sectional view of inserted bolt assembly used to connect the HDPE frame members to the galvanized steel frame corner units. The bolt assemblies are inserted from the outside surface of the HDPE frame members. Each bolt assembly consists of a 2.0-inch x 0.5-inch steel bolt with an outside flat washer, an inside lock washer, and an inside mounted nut. All hardware components are either galvanized or stainless steel.

OCAT MODELING AND DESIGN EVALUATION

A series of OCAT cage design evaluation and performance testing studies was conducted in 2007 at the Chase Ocean Engineering Laboratory facilities of the University of New Hampshire. Four studies were conducted under funding from the U.S. soybean industry:

• Numerical modeling to conduct simulation testing of the cage and mooring system under different sea states (USB Project 7512)

• Structural modeling to investigate cage framework stresses and failure modes under various loads USB Project 7513)

• Physical modeling to determine drag forces and test the impact of various wave configurations on a scale model of the OCAT cage (USB Project 7514)

• Design evaluation and construction of a modified OCAT cage based on recommended design improvements (USB Project 7515)

Summaries of the results of these studies are reported below.

NUMERICAL MODELING - OVERVIEW

In 2007, at the request of the U.S. soybean industry's OCAT project, engineers with the Department of Mechanical Engineering and the Center for Ocean Engineering, University of New Hampshire (UNH), constructed a numerical model of the OCAT cage to predict cage dynamics and mooring loads. A numerical model developed at UNH for modeling aquaculture systems, Aqua-FE, was used to predict the OCAT system behavior in the marine environment. Through the use of the Aqua-FE program, dynamic responses and corresponding forces acting on the OCAT cage and mooring system were investigated in a variety of wave and current regimes. Specific objectives were to: 1) predict heave, surge and pitch motion; 2) investigate the submergence depth of the cage in various current regimes; and 3) determine the tension in various mooring components.

A summary of the numeric modeling results is presented in this section. Complete details of the UNH numerical modeling study of the OCAT cage are available in the UNH final report to the U.S. soybean industry for USB Project 7512: Numerical Modeling of the OCAT Cage and Mooring System (Celikkol et al., 2007a).

NOTES ON THE CAGE AND MOORING SYSTEM

The single point mooring configuration, used by the OCAT system, allows the cage to line up with the prevailing current. The submergibility of the system in current depends upon the cage and float reserve buoyancies. The cage buoyancy is variable and the system's center of gravity can be adjusted by the location of the water in the various pipe segments (top, bottom or angled pipes). After an initial sensitivity study, it was found that the cage responds favorably with the following ballast configuration:

The resulting waterline rested approximately halfway up the top rim, providing 2470 N of buoyancy. This configuration was determined to be the most logistically feasible. Achieving this waterline in the field is straightforward and it will help insure that the cage has the proper buoyancy. In addition, the remaining buoyancy allows the cage to submerge in high current regimes. ENVIRONMENTAL (LOADING) CONDITIONS

Dynamic responses of the OCAT cage and mooring system were determined for regular and irregular waves, current profiles and a combination of both. Seven regular wave frequencies were investigated as well as two random sea events: a typical (operational) type event (Table 1) and an extreme (storm) event. Two typhoon type events were applied to the OCAT cage and mooring system. First, a JONSWAP wave spectrum was generated with a wave height of 5 meters and dominate period of 9.6 second (Table 2). A second load case was also applied to the numerical model which used the same wave regime but included a co-linear current of 1.5 m/s.

Table 1. Regular wave governing parameters.

The cage's response to increasing currents is essential to analyzing its single point mooring design as well as the effectiveness of the cage and mooring as a "self submerging" system. As very few environmental parameters for potential deployment sites of the OCAT cage system were known and because this cage was designed to be deployed in a variety of sea conditions, a range of currents was chosen. Nine different current profiles were applied ranging from 0-2 m/s (4 knots) (Table 3). Finally, a combination of waves and current was also simulated. In each load case, the mooring loads and cage dynamics were recorded.

Table 3. Input parameters for current testing.

RESULTS

Regular Wave Simulations

In general, the OCAT system showed an over-damped response at each wave frequency for regular wave simulations. The cage remained relatively stationary to the oncoming wave, letting the orbital wave motion pass through the cage frame while creating little additional motion. Wave Regimes 4 through 7 created the largest disturbance to the cage system, while Regimes 1 and 2 barely affected the motion of the system. Also, no resonance was found for any of the wave loadings.

Loads tested at four places along the lower mooring chain showed that loads at most points along the chain were relatively similar in magnitude. Wave Regimes 2, 3 and 5 created the largest loadings on the mooring and bridle lines. Tensions in Regime 2 reached upwards of 1500 N on the lower mooring chain, with over 800 N forcing to the upper bridle line pair. These load amplitude estimates, in conjunction with the expected range of tensions of certain mooring components, can help with future deployments.

Random Wave Simulations: Storm Conditions

The cage system was found to take a long time to reach a steady state under random wave loading. It was found that the system in heave does not become excited, and that the system in surge is relatively stable.

A storm event was also simulated with a collinear current of 1.5 m/s. This had a significant affect on the cage system's motion response. A highly damped response occurred because the cage submerged to a depth of roughly 15 meters (Table 4). At this depth the cage is no longer subjected to the full energy of each passing wave. In addition, the mooring line is stretched out with the current giving the cage slightly more stability with the constant tension.

Table 4. Steady state cage depth when subjected to random storm regime with current.

The random storm wave with applied current produced tensions an order of magnitude larger than those generated from the Random Storm Wave Regime without current. This aids in the explanation of the damped motion response for the cage with current. Also, similar to previous storm results, load responses at each mooring chain position match closely. It is important to note that for the operational storm spectra, wave heights, and thus energy, associated with the longer period waves was small.

Random Waves: Operational Condition

The motion response of the cage system in the Operational Wave Regime without applied current resulted in lesser magnitudes than those of the storm random wave regimes. Operational Random Wave motion responses for the cage system when subjected to an additional current of 0.148 m/s showed that motion is damped due to the application of current; however the current is smaller than that applied in the Random Storm Regimes so there is still some cage motion at the lower wave frequencies. For the lower mooring chain, the load responses at each of the four positions are very similar and relatively less than those loads from the storm regimes.

Simulations of Cage with Current Loading

Average tensions in the mooring and bridle lines when subjected to the eight current regimes are displayed in Table 5. The associated depth response of the cage is represented in Table 6. As expected, the tensions in each line increased with growing current velocities. Table 5. Average tensions in mooring and bridle lines in multiple current regimes.

Table 6. Depth response of the cage in 24 meter deep water with increasing current regimes.

The cage system remained at the surface for lower current regimes. With a 0.75m/s current velocity, the cage drops less than one meter. It is not until a 1.00m/s current is applied that the cage responds with any significant change in vertical displacement, reaching a depth of almost 8 meters.

At 2.00m/s, the lower bridle lines reached average loadings of 23 kN, while the lower mooring chain, nearest the anchor, experienced almost 60 kN of tension. These large tensions generated nearly 19 meters in vertical displacement. Originally located 4.5 meters below the sea's surface, the cage's horizontal bottom rims descended to a depth of less than one meter above the seafloor. At this depth the cage pendant chain was resting on the seafloor. This proves that the cage ballast not only adds stability and mass to the cage system, but provides the system relief in the event that the cage submerges to full depths. With the pendant chain resting on the seafloor, the reserve buoyancy of the cage increases, keeping the lower rims from potential damage when colliding with the seafloor.

CONCLUSIONS

The cage dynamics of the OCAT fish cage system were examined in this study. The cage design was tested with monochromatic waves, random seas and increasing current regimes. Motion responses of the system as well as associated mooring loadings were determined for each of the regimes.

The cage was found to have a highly damped response in heave. Waves tended to pass through and around the cage system and a resonant wave frequency was not found. The cage responded in surge and pitch as expected due to the lack of a resonant wave frequency. This provides evidence that the OCAT cage is sound for deployments at sites that experience similar environmental loading conditions. The cage was predicted to submerge under various current velocities, thus validating the "self submerging" feature. With the buoyancy distribution discussed (100% in lower and upper rim pipes and 55% in diagonal frame pipes), the cage did not come in contact with the seafloor, although the ballast chain did. The maximum tension at a current velocity of 4 knots was 60 kN, occurring in the lower anchor chain. This value was below the breaking strength of the mooring components. The maximum chain mooring load of 100 kN occurred during the storm wave simulation, with applied 1.5 m/s current, in the anchor chain. The maximum bridle line load occurred as a snap loading during the same storm event.

STRUCTURAL MODELING

OVERVIEW

In 2007, at the request of the U.S. soybean industry's OCAT project, engineers with the Department of Mechanical Engineering and the Center for Ocean Engineering, University of New Hampshire (UNH), prepared a structural model of the OCAT cage system to evaluate its robustness. A standard finite element (FEA) analysis software package, MSCMARC Mentat, was employed to analyze the cage frame. FEA was utilized to examine the stress concentrations, possible failure of components and potential areas of failure.

The following summary of the UNH structural modeling study of the OCAT cage is taken from the UNH final report to the U.S. soybean industry for USB Project 7513: Structural Modeling of the OCAT Cage (Celikkol et al., 2007b). That report describes the approach used to analyze the components and the mechanical properties of HDPE, the two models developed, boundary conditions, and results.

METHODOLOGY

Structural model testing was conducted through the development of a three dimensional finite element model of the OCAT cage frame to predict structural integrity of the cage. This model was developed in the commercially available FEA software package, MSC. Marc®. Numerical simulations were performed for mechanical loading to analyze stress concentrations, possible failure of components and potential areas of failure. Load cases for the structural OCAT cage frame were taken directly from the output load forces identified in the Numerical Modeling study (USB Project 7512) for multiple wave and current simulations, specifically the 2 m/s current and the irregular storm wave conditions with 1.5 m/s current. Two types of finite element analysis models were constructed for this investigation, one of the entire cage frame using beam elements, and one of the most loaded rim section using shell elements.

CAGE FRAME ANALYSIS RESULTS

The numerical simulations provided predictions on the deformation of the cage frame, distribution of the bending and twisting moments and axial stresses in the frame. The deformed shaped produced is presented in Figure 1. The results of the simulations were processed to determine the cage frame component which experienced the maximum loading. For all of the structural analysis it was assumed that the steel corner fittings were over-designed for the model; subsequently it was expected that the HPDE rim sections would fail first. This assumption was further discussed and verified in the section Modified OCAT Design.

Figurel. The deformed cage frame. The lighter purple outline is the undeformed shape. The results are magnified 3 times for clarity.

Typically, bending moments contribute the majority of the forcing within an element, therefore the frame was analyzed to find the HDPE rim with the largest total bending moment. Once the HDPE rim sections with the largest bending moments were identified, their resulting total bending moment, torsional moment and axial force were recorded for the detailed testing. These results are depicted in Table 1. Results showed that the diagonal frame members experienced the highest bending moments (load case 1), specifically the upper most sections (closest to the upper rim). For the distributed force (load case 2), the lower horizontal rim experienced the highest loads.

Table 1. Point and distributed frame anal sis results for a lied current and storm loadings.

As expected, the largest loadings were generated in the random storm conditions due to snap loads in the bridle lines. Total bending moments, as well as axial forces, were larger in the distributed load analyses than those of the point loads. The axial forces were significantly larger in magnitude than those found from the point load analyses. This is due to the position of the pipe sections. The horizontal rims are nearly parallel to the direction of fluid flow, or loading vector, so the largest forcing would expectedly be in the same direction (in this case, the axial direction). In the point load cases, forces were for the most part not directed along the diagonal rim axis, but instead pulled out and down, producing large bending moments.

DETAILED RIM ANALYSIS

The results for the cage frame analysis were used as input to the detailed rim section model. The two most common failure mechanisms, material failure and localized buckling of a pipe, were investigated. To assess material failure, four simulations were performed to investigate the areas of highest stress concentrations in HDPE generated from the specified loading to determine if the pipe suffered from material failure. The diagonal and lower horizontal rim were both investigated with the results from the current and storm loading on the cage frame. To assess localized buckling potential, a set of simulations was performed to determine if and when the HDPE would buckle in each load case. Both the diagonal and horizontal detailed rim sections were loaded under the current and storm forces to determine when localized buckling occurred.

DETAILED RIM ANALYSIS RESULTS

Material Failure

The rim shell model was subjected to loading moments and bending, torsional and axial forces to determine the highest stress concentrations and see if material failure would occur. The largest stresses were produced by storm conditions. This was expected since large snap loadings in the bridle line created forces of significant magnitude. Similarly, larger bending moments from the distributed frame analyses applied to the horizontal rim sections generated larger stress areas than those from most of the diagonal simulations. One interesting result was that the bending moment applied the majority of the stress in each component. The addition of the torsional moment and axial forces only increased the stress in the diagonal and horizontal rim sections 1.5% and 10%, respectively. It was found that from the cage frame analysis that the torsional moment values were very similar between the two rims; however, axial forces applied to the horizontal rim far exceeded those to the diagonal rim. Although the horizontal rim is larger and thicker than the diagonal rim, this results in a noticeable increase of equivalent stress.

The largest Von Mises stress, 1.033 x 10⁷ Pa, occurred in the horizontal rim under the storm wave loading. However, this value is more than two times smaller than the published yield stress value of HDPE, 2.413 x 10⁷ Pa. Figure 2 illustrates the deformed horizontal rim for this loading scenario. Therefore, it can be assumed that the HDPE pipe does not suffer from material failure.

Figure 2. Slightly exaggerated view of horizontal rim' s deformed shape with storm distributed loading conditions. Color bar illustrates Von Mises Stress values in the outer layer. Localized Buckling

The second set of shell analysis simulations involved increased loadings (100 times) to obtain localized buckling as a mode of pipe failure. The model was monitored until pipe failure occurred. Table 2 displays the final moments and forces that generated localized buckling. Buckling was assumed to occur as half a nodal circumference reached the yield stress.

Table 2. Localized buckling analysis: predictions for critical loading.

It was found that the applied bending moment is the largest contributing factor to potential localized buckling. The minimum bending loading at which failure to the diagonal rim was initiated occurred in current loads with a bending moment of 16.50 kN-m. The minimum bending loading at which failure to the horizontal rim occurred in storm loads with a bending moment of 45.35 kN-m. Figure 3 illustrates this diagonal rim model just after localized buckling occurs. The HDPE rim seems to bulge on one side as it is stretched away from its attachment to the steel fitting on the opposite side.

Figure 3. Diagonal rim member subjected to loading just after localized buckling occurred.

Safety Factor Calculation

Each rim member with the largest applied bending moment was then studied to determine the component safety factors. Both the diagonal and horizontal rims' maximum bending moments occurred in simulations with storm applied conditions (as found in the previous section). By comparing these values, the factor of safety for each component was calculated.

It was important to make sure that not only the pipe member being analyzed did not fail, but the other components of the cage frame were also not failing. For example, if the diagonal rim member was being tested for localized buckling under a specified loading condition, it was necessary to investigate the maximum bending moments occurring in the alternate rim member, in this case the worst loaded horizontal rim member, to ensure no member was failing under the same loading. These moments are displayed in Table 3. With these values, safety factors were calculated for both rim members in the two worst loading scenarios. The results for all safety factors are shown in Table 4.

Table 3. Maximum a lied bendin moments for both rim members.

Table 4. Safety factors for each member's maximum applied bending moment, as well as their alternate rim' s safety factor for the same simulation.

Safety factors show that in both worst case loading scenarios, the respective rim member reached 25% of its failure loading. The alternate rim member in both cases did not exhibit a lower safety factor. Therefore, under the selected load cases, the pipes did not suffer from localized buckling. It is also important to note that these values are conservative, due to loads applied during the storm were snap loads. These loads occur over a very short time frame, where HDPE will be more resistant to deformation and failure.

CONCLUSIONS

The structural integrity of the OCAT cage's rigid frame design was investigated in this study. The results from two environmental load cases were used as input: 2.0 m/s current velocity and storm wave regime with 1.5 m/s current. The entire cage frame was first analyzed using beam analysis with two sets of load cases. These load cases bracket the stresses in the cage frame in a marine environment. A conservative value of the HDPE modulus of elasticity was also used. A second analysis investigated structural reliability of the most loaded rim sections.

The frame analysis found that storm conditions applied the greatest forces to the lower horizontal pipe and the diagonal rim sections. These forces were then used for a detailed rim analysis of the most loaded parts of the rim. The bending, axial and torsional forces were applied to the horizontal and diagonal rims over a series of simulations.

It was found that neither material failure nor localized buckling would occur under the considered load cases. In addition, minimum safety factors of 4 were predicted throughout the cage frame. In normal conditions, the safety factor will be much larger due to the severity of the loads applied. Therefore, it can be assumed that the OCAT cage system will survive the structural loading associated with the environmental conditions tested in USB Project 7512 - Numerical Modeling of the OCAT Cage and Mooring System.

PHYSICAL MODELING

OVERVIEW

In 2007, the U.S. soybean industry contracted the Department of Mechanical Engineering and the Center for Ocean Engineering, University of New Hampshire (UNH), to construct a scale model of the OCAT ocean cage system to evaluate OCAT cage reaction in different ocean wave and current regimes. A scale model of the Ocean Cage Aquaculture Technology (OCAT) system was constructed and tested in the UNH wave/tow tank. Tank testing was performed by subjecting the cage and mooring system to a series of regular and irregular waves so the mooring line loads and cage dynamic response could be measured. In addition, the cage was towed through the water to simulate current and the drag force was recorded. The objectives of the study were to determine the cage dynamics and mooring loads under waves and current. In addition, a series of tests were performed on a prototype fairing device (current blockage system) to ease towing of the cage or to allow deployment in high water velocity areas.

The following summary of the UNH physical modeling study of the OCAT cage and mooring system is taken from the UNH final report to the U.S. soybean industry for USB Project 7514: Physical Modeling of the OCAT Cage and Mooring System (Celikkol et al., 2007c).

CAGE CONSTRUCTION

The OCAT fish cage and mooring system was modeled with a scale factor of 1: 10 and tested in 2.4 meters of water (24 meters full scale) in the UNH wave/tow tank. The scale cage and mooring are shown in Figure 1.

Figure 1. The 1:10 scale OCAT cage (a) and mooring (b). The cage was secured in the mooring with a 4-point bridle, chain, float and deadweight (not shown).

EXPERIMENTAL DESIGN

Physical model testing was performed in the Jere A. Chase Ocean Engineering Laboratory's wave/tow tank facility at the University of New Hampshire. The tank is 36.5 m (120 ft) long, 3.05 m (10ft) wide and 2.44 m (8 ft) deep. The flap-type wave maker is capable of generating monochromatic and random waves. A carriage system, resting on top of the tank, allows for model towing.

The OCAT cage scale model was tested in both waves and current. The OCAT cage and mooring system were subjected to regular and irregular waves and current. Seven regular wave frequencies were investigated as well as two random sea events: an extreme (storm) and typical (operational) type event. Eight different current speeds were applied ranging from 0 - 2 m/s (4 knots) full scale. Regular Wave Regimes

A total of seven regular wave regimes were tested as shown in Table 1. Each simulation was run for a total of 30 seconds model scale (roughly 95 seconds full scale), though not all simulations were recorded for this entire length. The last two wave regimes, 6 and 7, were only run long enough to fully capture the first two waves due to tank limitations.

Table 1. Regular wave input parameters for physical scaled model testing.

A typhoon type event was applied to the OCAT cage and mooring system. A JONSWAP wave spectrum was initially generated with a full scale wave height and dominate period of 5 meters and 9.6 seconds, respectively. However, the scaled significant wave height could not be generated in the wave tank. A maximum allowable significant wave height of 0.13m (1.3meter full scale) was used. It can be expected that this would affect the motion response of the cage. Also, unlike the numerical modeling study (USB Project 7512), only non-current random seas could be tested in the wave tank.

Random Waves: Operational Condition

A second set of random sea analyses was completed to better understand the dynamics of the cage system on an average (or operational) day. This scenario was tested so field personnel can predict the system response on a day when they would typically travel to the site. Using similar methods as the storm event, a random sea with a full scale 1.0 meter dominant wave height and 5.34 second dominant period was generated (waves with a 5.34 second period are typical wind driven waves). Table 2 presents the tested JONSWAP parameters. Table 2. Random wave input parameters for physical scaled model testing.

Current Regimes

The cage's response to increasing currents is essential to analyzing how the cage will respond. This information is also useful for the design and testing of a fairing device, discussed later in section Tow and Current Deflection Fairing. A series of water velocities, increasing to 2 m/s of current full scale, were applied to the system (Table 3). To ensure acceptable results, each current regime was replicated three times.

Table 3. Current velocity input parameters for physical scaled model testing.

RESULTS

Regular Wave Tests

OCAT cage heave, surge and pitch motion responses to regular waves were identified from wave tank tests. It was found that wave regimes 4 through 7 generated a wave contouring behavior, where the cage closely followed each oncoming wave with little to no damping or resonance. Contrarily, Wave Regimes 1 and 2 were dominated by the cage system inertia, allowing waves to pass over and through while creating little motion disturbance to the cage. Finally, Wave Regime 3 reached a middle ground responding to 50% of the actual wave height.

In general, the motion response of the cage system seemed to vary little with the addition of the weighted submersible load cell in the lower mooring chain. The largest differences were consistently found in the pitch response of the cage. Heave and surge responses fluctuated with less than 10% for most regimes. Wave Regimes 2 and 3 experienced the largest maximum tension on the mooring chain at any given point during interaction with the specified wave field, reaching magnitudes nearing 2 kN loading, full scale (Table 4).

Table 4. Full scale maximum lower mooring chain tensions for the regular wave regimes.

Random Wave Tests

The random storm wave regime reached a maximum tension of a much higher magnitude than that of the random operational wave regime, as expected. The motion response trends for the Random Storm and Operational Waves were similar to the results form the regular wave testing, and seemed to vary little between the tests with and without the load cells.

Maximum mooring tensions, obtained from a submersible load cell located in the lower mooring chain nearest the upper mooring chain connection, were also of a much higher magnitude for the random storm wave regime than for the random operational wave regime. Maximum mooring chain tensions are reported in Table 5.

Table 5. Maximum lower mooring chain tensions for the random wave regimes. Note: no current applied.

Tow Tests

The OCAT cage was also tested at eight water velocities, and the system drag was measured. The steady state portion of the tow was used to determine the average drag of the OCAT system. Figure 2 displays full scale average drag force at each velocity. The highest water velocity of 2.0m/s (4 knots) subjected the bridle line intersection to forces over 40 kN. The cage is shown towed at low and high water velocities in Figure 3.

Figure 2. Full scale average drag forces subjected to the cage's bridle intersection with corresponding average velocities for each of the current regimes.

Figure 3. The OCAT cage under tow. The cage and tow staff provided virtually no disturbance at low water velocities (a). At high water velocities, the opposite was true (b). TOW AND CURRENT DEFLECTION FAIRING

One objective of the physical model experiments was the design and analysis of a system to be used for cage towing applications or deployment in high water velocity environments. Having a method to tow the OCAT cage has advantages for situations in which a live harvest has economic and logistical advantages. Harvesting fish from the cage could be performed by attaching the cage to a boat and towing it to shore for unloading. When towing the cage, it must remain submerged and the interior calm to keep the fish alive during transport. Live fish have a much higher market value then dead fish, and it is important that the fish are not stressed during transport. The cage must have relatively low drag in order to accommodate the tow boat. In addition, some of the same methods used for towing can be used when the OCAT is moored in areas of high current.

The OCAT cage drag force as a function of relative velocity was investigated and discussed in the previous sections using a 1/10 scale physical model in Froude-scaled, tow tank experiments. Since both storm currents and tow speeds could be up to 2 m/s (4 knots), shielding the fish by introducing external fairing may be advantageous. Specifically, the addition of flexible fairing to the cage has potential for reducing the drag force as well as protecting the fish from the incident velocity of water. Fairings made of durable fabric and could be fastened directly to the cage. This would make the attachment in the field feasible and not require heavy infrastructure. UNH developed five different fairing design alternatives to implement on the offshore cage.

1. Design 1 - Attachment of a fairing onto the front cross section of the cage and using the existing four point bridle to tow the cage (Figure 4).

2. Design 2 - Attachment of the same fairing on the front as in Design 1 ; however, an additional fairing was added to the base of the cage. The two bridle lines attached to the top of the cage were removed, and the cage was towed using just the bottom bridle rope linkages (Figure 5).

3. Design 3 - Attachment of two side panel fairings were adjacent to one another. A new two-point tow bridle was created (one side of the original four-point bridle), which attached to the lower and upper corners of the cage across the common edge of the two fairings (Figure 6).

4. Design 4 - Attachment of a cone fairing to the top of the cage. The cone extended from the top of the cage to a point where a single point towline was attached (Figure 7).

5. Design 5 - Attachment of two fairing panels to the side edges of the front of the cage using the concept of a plow fairing. The panels extended outwards in front and met along a common edge. A bridle line was attached to the top and bottom of the leading edge, creating a two-point harness to tow the cage (Figure 8).

Figure 4. Design 1 with single front panel towed with the four-point mooring bridle.

Figure 5. Design 2 utilizing front and bottom fairing panels.

Figure 6. Design 3 utilizing 2 side panels and a 2 point harness.

Figure 7. Design 4 utilizing a cone fairing on top with a single point harness for towing.

Figure 8. Design 5 utilizing a 'plow' like fairing with a two-point harness.

To test each fairing idea accurately, multiple tests at different speeds were taken and averaged The cage was towed at 8 different speeds including 8, 16, 24, 32, 40, 48, 56, 64 cm/s (model scale) and repeated three times Drag force data was obtained for each design that showed initial promise after visual observation in motion If the design did not perform in the desired way after several tests, the testing of that design was stopped Once a design was determined to be plausible, a current meter was placed in various locations of the cage to measure the velocity reduction Using visual observations and measured data, a final conclusion on the best design was made Measurements at model scale were Froude-scaled to full size values

FAIRING TEST RESULTS

The initial thought that a fairing could reduce the drag of the cage was reversed after testing due to the increased drag with every design Figure 9 shows the drag force for each design on a common scale The lowest drag force for each velocity came from Designs 2 and 4 Design 2 had a better velocity reduction, but with the two- point bridle rolled without recovery at higher speeds (>32 cm/s) Design 4 had a lower drag force and consistently towed very smoothly at all speeds with the bottom edge riding along the surface of the water

Figure 9. Drag force of all designs vs. incident velocity. FAIRING EVALUATION CONCLUSIONS

Fairing the OCAT system can provide a positive addition to its design and functionality. The use of fairing can reduce the water velocity inside the cage to protect the fish during transport. Fairing Designs 2 and 4 indicate that the velocity can be reduced dramatically from the incident to provide sufficient protection for fish during cage towing. Design 2 offered complete protection throughout the cage at all the velocities tested, yielding slightly negative values. Design 4 offered complete velocity reduction at low speeds throughout the entire cage. As the velocity increased, the center of the cage remained protected, while velocities at the outer edges were reduced by approximately 25 %. Based on the tow test results, UNH recommended field testing Designs 2 and 4. Each configuration can be easily attached in the field, with no diving required in Design 4. At sea testing of the recommended fairings on the OCAT cage is scheduled to be conducted by UNH in 2008.

OCAT DESIGN EVALUATION AND MODIFIED CAGE CONSTRUCTION DESIGN REVIEW

In 2007, at the request of the U.S. soybean industry's OCAT project, engineers with the Department of Mechanical Engineering and the Center for Ocean Engineering, University of New Hampshire (UNH), conducted a design review of the original OCAT cage. This evaluation included a review of all component parts, identification of the availability and connectivity of component parts, assessment of fabrication and assembly costs, and identification of improvement options.

The UNH design review identified two areas for possible improvement: modification of the existing frame corner fittings, and/or removal of the steel corner fittings on one or both HDPE frame rims. A modified OCAT cage was constructed based on the identified improvements. The following details of the UNH design review and construction of a modified cage are taken from the UNH final report to the U.S. soybean industry for USB Project 7515: Construction and Deployment of an OCAT Cage in the Gulf of Maine (Celikkol et al., 2007d).

CORNER FITTING MODIFICATIONS

The original corner weldments, shown in Figure 1, secure the HDPE pipe and form the rigid cage frame. HDPE pipe is slid over the tubular steel extensions and through bolted. However, to make the HDPE-steel interface work properly, standard steel pipe could not be used due to dimension conflicts. Steel tubing was therefore utilized and machined to fit. Tubing is typically harder to obtain and more expensive than steel pipe. When the machining charge is factored in, the pieces become even more costly.

Figure 1: Diagrams of the original OCAT corner weldments. Steel tubing is utilized for the majority of the corner pieces, increasing the material and fabrication cost.

The original OCAT corner units weighed 105 pounds (47.6 kg), which was beneficial as they could be easily moved. However, upon investigating the fabrication price, UNH found that each fitting would cost approximately US$3,250, resulting in a total cost of US$26,000 for the eight corner units needed for the cage frame. This price was a direct result of the steel tubing and custom nature of the fittings.

As a result of the projected cost, the corner fittings went through a series of design modifications where the fittings were simplified and constructed from readily available components. First, the diagonal pipes in the cage frame were increased to the same diameter as the lower and upper rims. Then 10-inch (25.4-mm) standard steel pipe fittings and flanges were incorporated into the corner design. Unlike in the original design, however, the HDPE pipe would not slide over steel tubing extensions, but bolt together using flanges (Figure 2).

A prototype modified corner unit is shown in Figure 3. The modified corner unit is constructed of standard 10-inch nominal steel pipe (O.D. 10.75 inches), and is designed to mate with 10-inch DR 11 HDPE pipe (O.D. 10.75 inches). Each modified corner unit consists of one standard 90-degree tee, two standard 45-degree elbows, and three steel flange units with one each welded to the tee and the two elbow pieces. These parts are readily available and shipped prepped for welding. The 90-degree tee and the 45-degree elbows are welded together so that the tee is correctly angled for the diagonal HDPE pipe. The general layout of the mooring and pad-eyes remained the same as in the original design, although the plate dimensions were increased to handle the larger steel pipe.

Figure 2. Schematic drawings of the new corner fitting design. The corners are made of standard 10- inch pipe fittings (A). The HDPE pipe segments will have a stub end fused on, and a backing ring can be used to bolt the components together (B).

Figure 3. Photo of the modified OCAT cage corner unit. The modified corner unit is made from standard 10-inch steel, and consists of one tee and two 45° elbow units welded together, and with welded steel flanges on the tee and elbow units to bolt the horizontal and diagonal HDPE pipe frame segments to the steel corner units.

The rim and diagonal frame members are constructed of 10-inch SD 11 HDPE pipe. The HDPE pipe has an O. D. of 10.75 inches, an LD. of 8.72 inches, and an average wall thickness of 1.015 inches. The HDPE pipe segments had to be modified to face and attach to the new fittings. To do this, two-piece flanges, consisting of an HDPE stub end and a steel backing ring, were fused onto each end of the HDPE pipe (Figure 4). Once attached, the steel backing ring is bolted to the corresponding steel flange welded on the corner unit tee and elbow segments (Figure 4).

Figure 4 (right). An HDPE flange is welded to the end of each HDPE pipe segment of the OCAT cage frame. A 10-inch I.D. steel flange is then slipped over the pipe and butted against the HDPE flange. This flange is bolted to a matching flange on the steel corner unit to connect the corner unit and HDPE frame member.

These modifications simplify the corner fabrication, reduce the overall cost and simplify the cage construction. Table 1 lists the quote for the original fittings and the final cost of the new design. The new corners should reduce the cost of the corner fittings by over 50%. Cage construction is simplified in that the HDPE pipes no longer have to be forced onto steel tubing. Bolt holes no longer have to be drilled in the HDPE pipe segments. The steel corner units and HDPE pipe segments are aligned by the rotating backing rings. In addition, deconstruction and reassembly becomes less tedious because of the new modularity of the system.

Table 1. Material and fabrication costs for one corner fitting.

It is important to note that the weight of the new corners increased to 338 lbs (153.3 kg). This additional weight can complicate moving these modified parts. To resolve this, corner unit supports were fabricated. These supports were designed to have wheels for movement as well as support the entire weight of the cage if necessary. The supports are shown in Figure 5. One unexpected benefit of the supports, determined during construction, was the cage rested off the ground, easing assembly. Drawings for the new corners and supports can be found in Appendix A and B, respectively.

Figure 5. Photo of the rolling fitting supports for the cage corner units. The supports cradle the corner fitting as well as provide a means of easily moving the fittings. FULL HDPE RIMS

A second observation made during the UNH design review was to construct one or both of the horizontal rims entirely out of HDPE pipe and fittings. Eliminating the steel fittings on one rim would possibly: (1) reduce the cost of the cage system; (2) decrease the weight of the cage; and (3) determine the feasibility of constructing a rim entirely out of HDPE.

A rim made entirely of HDPE can be constructed in two ways: 1) fabricate four HDPE corner fittings, similar to the new fittings discussed in the previous section, and bolt these fittings and pipe segments together to form a rim; or 2) fuse the fittings around the circumference, thereby eliminating bolted connections. Although utilizing bolted HDPE corner fittings has an added benefit of modularity, it was decided that the top horizontal rim would be constructed by fusing the HDPE into a continuous ring, as depicted in Figure 6. The lower cage frame would still incorporate the modified fittings discussed in the previous section, however the upper rim would be assembled as one piece. If the construction proved to be difficult, HDPE or steel corner fittings would be manufactured and utilized. To assemble the top rim, 10-inch HDPE fittings were used: one 90° tee, two 45° elbows and sections of HDPE pipe cut to the proper lengths.

Figure 6. Representation of the proposed single piece HDPE cage top rim. The top rim is constructed of fused of HDPE fittings. HDPE stud ends for connection to the diagonal HDPE frame pipes are attached via steel backing rings (not shown).

IMPACT OF THE MODIFICATONS

The two major differences in the OCAT cage constructed by UNH and the original OCAT design were the new steel corner fittings and assembly of one rim constructed entirely of HDPE. These modifications change the system's center of gravity (CG), center of buoyancy (CB) and the transportation water line. Comparisons between the original and modified OCAT cages are shown in Table 2. The new system's CG is lowered by 0.51 m, thus making it more stable. However, the cage will also sit lower in the water. The original cage design rode on top of the water during towing operations. Due to the increased weight, the new system has a draft of 1.6 m. This can be reduced by adding temporary floats for the tow that can be removed once at the site.

Table 2. Hydrostatic comparisons between the original and new OCAT frame designs.

Finally, because the top rim was constructed of HDPE, it will be 45° out of phase with the lower rim (Figure 7). However, this is expected to have minimal influence on the response of the cage in waves and current.

Figure 7. Top view of the modified OCAT cage frame design. The design of the fused, single piece HDPE top rim is 45° out of phase with the lower rim.

CONSTRUCTION OF THE MODIFIED OCAT CAGE

Construction of the modified OCAT cage frame was performed in three stages: 1) fitting fabrication, 2) rim/pipe fusing, and 3) system assembly. The following sections describe each stage.

Corner Fitting Fabrication

The new corner fittings were constructed from standard 10 inch pipe fittings: one 90° tee, two 45° elbows and three weld neck flanges. The pipe fittings were purchased and delivered to a commercial workshop for corner fabrication. There the parts were rigged, the mooring and ballast pad-eyes were machined, and the components were welded together. Upon completion, the fittings were painted and prepped for assembly (Figure 8). The fitting supports were also fabricated at the commercial workshop.

Figure 8. Newly painted corner fittings. Eight modified OCAT corner fittings were fabricated for the new OCAT cage. Two mooring pad-eyes were welded onto the outside face (one shown in right photo). The ballast pay-eye was placed on the inside portion of the fitting.

Rim/Pipe Fusing

Step two in the construction process was fusing of the two-part flanges onto the HDPE pipe segments, followed by assembly of the top rim. A #412 fusion machine, shown in Figure 9, was rented from an HDPE pipe installation company. Using this device, the pipe segments and top rim were joined.

Figure 9. Photo of the #412 fusion machine used for rim and pipe fusing.

The lower and diagonal pipes were first cut to length. Then the stub end and back rings were fused on each end (Figures 10 and 11). Each fusing required approximately 15 minutes. Complete fusing of the pipe segments required 8 man-hours of labor. In parallel to the fusion process, the pipes were plugged to form individual air chambers.

Figure 10. Photo of the lower and diagonal pipe segments. A stud end and steel backing ring was fused onto each end of the HDPE pipe.

Figure 11: Each pipe segment was sealed with welded HDPE plugs.

The top rim was fabricated in a similar manner. Using the correct assortment of 90° tees, 45° elbows and short lengths of HDPE pipe, the top rim was fused to form a single rim. Special care was needed when fusing the 45° elbows to the 90° tee to insure that the extension portion of the tee (the portion that connects to the diagonal pipes) was attached at the correct angle of 52°. As the upper rim was fused, it was necessary to use a forklift or boom truck to support and sometimes move the rim as it grew in length. Once the rim was complete, the handrail and splashboards were welded on (as seen in Figure 12). The formation of upper rim took approximately 16 man-hours.

Figure 12: The upper frame assembled with the handrails and splashboards. OCAT System Assembly

Once the fittings and pipe segments were complete, the OCAT system was prepped for assembly. All components were shipped to the New Hampshire Port Authority in Portsmouth, NH. The first step was to construct the lower rim. The fittings and lower rim pipe were first laid out, as shown in Figure 13. Gaskets were then placed against the flanges and the HDPE pipe segments were temporarily bolted to each fitting to ensure proper orientation. Upon inspecting the lower rim to verify it was square, the remaining bolts were installed and tightened. A lubricating agent was applied to the bolt to avoid seizing in the marine environment.

Figure 13. Photos of the lower rim assembly. First the pipes and fittings were layout out. Then the pipe was temporarily held in place to ensure the lower frame was square. Finally, the remaining bolts were installed.

Once the lower rim was assembled, work on the top rim and diagonal pipes began. To do this a crane, a boom truck and a scissor lift were employed. The diagonal pipes were first attached to the upper rim and then the top framework was connected to the lower rim. A small crane suspended the top rim while the boom truck supported one diagonal pipe. Then, using the scissor lift, personnel bolted the first pipe to the upper rim. This process was repeated until all four legs were attached (see Figures 14 through 16). The crane then suspended the upper superstructure over the lower rim and the bottom connections were made.

Figure 14. The diagonal pipes were attached to the top rim by a crane, boom truck and scissor lift. The boom truck suspended the upper rim while the crane lifted the diagonal pipe sections into place. Personnel in the scissor lift bolted the connections as each diagonal segment was lifted into position.

Figure 15. Photo of the diagonal pipes being bolted to the upper rim.

Figure 16. Photo of the top structure being lifted over the lower frame and attached.

The next step was the addition of the air system. The air system was needed to reduce scuba diving dependence for the planned tests in USB Project 7516 - Submersion Testing of the OCAT Cage. Holes were drilled and tapped into the lower and diagonal HDPE pipes for the intake/exhaust of air/water (Figure 17). Brass inlet valves and fittings were then fitted. All air inlet valves and fittings were 3/8 inch. For the water inlet valves on the angled and bottom pipes, 0.75-inch valves were used to speed up the intaking and exhausting of the water in the pipes. Thread sealant (liquid Teflon) was used on the valve threads to ensure that air will not leak by the valve. Experience has shown that this method of drilling and tapping HDPE works well.

Hoses were then run from these valves to the upper rim to allow the cage to be raised or lowered from the surface. Standard multipurpose air hose was used for the air lines. These were installed for the short test in the estuary, but are not recommended for an extended time in saltwater. Air lines are available that can be used that are wear resistant. These typically have a metal sleeve woven into the outer skin to prevent wear.

The recommended option is a combination of standard air hose and quick connect couplings (Figure 17). The simplicity of the OCAT cage is its biggest asset. By being able to quickly add or remove air hoses, there is no need to purchase more expensive wear resistant hose, or to make sure that the hoses are securely attached to the cage frame. Having easily connect/disconnect hoses allows for control of the cage buoyancy from the surface, and only requires quick dives to attach and remove the hoses when buoyancy corrections are required. These can also be used on the original OCAT design.

Figure 17. Drawing of a quick-connect coupling to attach air hoses to air and water intake/exhaust valves on the OCAT cage frame.

Once the air system was installed, the net was secured within the cage framework. The net was attached to eyebolts, placed in the flanges of the diagonal pipes, with stainless steel clips having a locking mechanism (Figures 18, 19 and 20).

Figure 18. Photo of a lower corner fitting showing the lower rim air valve for intake and exhaust of air and water. The net was attached to an eye-bolt, which replaced one of the twelve standard bolts in the flange frame. Air was exhausted from the lower rim via a valve and hose leading to the surface.

Figure 19. The net was attached to the frame with stainless steel clips with locking sleeves. The clip is rated to 3500 lbf. There is a threaded sleeve (screw lock gate) which locks the clip in the closed position. The screw-lock gates open easily with one hand.

Figure 20. Photo of the net connected to the flange eye-bolt with a stainless steel, locking sleeve clip.

Following the net installation, the ballast bridles and chain were attached to the pad-eyes on the lower corner fittings (Figures 21 and 22). The ballast bridles were cut to length and two thimbles spliced in each end. One-half inch shackles were used to secure the ballast bridles to the lower fittings via the "inside" pad-eye. The ballast bridles were then secured to the ballast chain with 0.5 -inch shackles and a steel link.

Figure 21. The ballast bridles contain thimbles spliced into each end and are attached to the pad-eyes on the lower corner fittings with a 0.5-inch galvanized or stainless steel shackle.

Figure 22. The ballast bridles are secured to the ballast chain with 0.5- inch shackles and a steel link.

The upper and lower bridle lines were then attached to the cage frame. The lower bridle lines were spliced with a thimble. They were then shackled to the lower fittings (shown in Figure 23) via the pad-eyes similar to the original specifications.

Figure 23. The lower bridle ropes were shackled to the pad-eyes in the lower corner fittings.

Because the upper rim was made as a single unit, shackles could not be used. The upper bridles were tied to the upper rim, at the locations shown in Figure 24, similar to how salmon net pens are secured. A clove hitch was tied around the rim. An additional half hitch was tied with the tail end of the line to prevent the knot from untying.

HDPE pad-eyes were investigated for the upper rim and could be fabricated and welded to the HDPE corners. However, to ensure that the HDPE weld would be sufficient, a lot of weld area is needed and this was considered too risky for the initial study. In the future, a collar with a built in pad-eye could be employed for bridle attachments (Figure 25)

Figure 24. The upper bridle ropes are tied to the upper rim at the locations shown.

Figure 25. A collar with a built in pad-eye could be employed for attachment of the upper bridles to the HDPE upper frame rim.

Finally, instruments for field testing were installed An Aquadopp current meter was placed in the center of the net chamber to measure any water velocity reduction Three pressure sensors were also added at the lower corners of the net to measure the submergence depth of the cage The instrument set-up can be seen in Figure 26

Figure 26. The modified OCAT cage was outfitted with three pressure sensors and a current meter.

The modified OCAT cage assembly, including cage frame assembly, air system installation and net attachment, took 32 man-hours (not including crane and boom truck personnel). The upper rim fabrication required 16 man- hours.

SUMMARY AND OTHER DESIGN RECOMMENDATIONS

Based on the construction experience, the modifications made to the cage were considered a success. The new fittings allowed the cage to be constructed with little difficulty. It was also proven that it would be feasible to construct an entire rim out of HDPE.

There is, however, room for future improvement. First, the cage frame should be altered to have the diagonal pipes rise at a 45° angle, instead of 52°. This would simplify the fitting fabrication, as the fusion process in assembling the one-piece upper rim was complicated. However if care is taken, the fittings are not overly difficult to make. Regardless, if the angle of the cage is changed to 45°, it will result in the following modifications. Assuming that the top rim size and cage net volume remain the same, i.e. approximately 2 meters in length and 100-m³ in volume, the lengths of the diagonal and lower rim frame pipes are increased to 6 m and 7.25 m, respectively. This then results in a slightly decreased overall cage depth of 4.25 m. Dimensions of the original specification, the assembled UNH cage, and the new modified lengths due to the 45° fitting angle are shown in Figure 26.

The modified dimensions will have little to no affect on the cage's structural integrity. The benefit are the reduction of any mistakes made during fitting fabrication, particularly if HDPE fittings are to be used, due to the 52° and 38° angle relationships, and easier calculation of the pipe lengths if any modifications are made during construction.

In the future, manufacturing the top rim with HDPE fittings, bolted together, may also be easier. One final note is that the ballast bridle line lengths should be increased. This would not only provide a greater restoring moment to the cage, but ease the bridle attachment to the chain ballast.

STRUCTURAL INTEGRITY OF MODIFIED OCAT DESIGN

Stress analyses of both the original and the new fittings (Figure 27) were conducted by UNH to evaluate the stress concentrations and determine any areas of concern in the corner fittings, and to ensure that the steel corners (original and new) would not fail under load. The fittings were modeled in MSCMARC Mentat using the forces previously described in this report. The stress concentrations and safety factors for both fitting configurations were found.

Figure 27. The original (a) and modified (b) OCAT cage corner fittings.

In an effort to model the worst case condition, the assumed "weakest" portions of the fittings were loaded. In the original design, the diagonal pipe attachment has a smaller diameter and thickness, yet according to the overall cage frame analysis, experiences a very large load. Therefore, this segment of both fittings was chosen for evaluation.

Analysis and Results

The simulations were run with linearly elastic analysis and the equivalent Von Misses stress was recorded. Von Misses stress is a standard criterion used for predicting yielding of ductile materials.

The results from the original and modified fitting models can be seen in Figures 28 and 29, respectively. The stress was found to concentrate at the joint of the diagonal and horizontal pipes on both fittings. The maximum stress for the original and new corner fittings was 1.8 x 10⁸ Pa and 3.2 x 10⁷ Pa, respectively. This represents an 80% reduction in the new corner weldments. This decrease in stress is a result of the larger pipe diameter and increased wall thickness in the new fittings. Also, the modified fittings have a smoother geometry, eliminating stress points (sometimes caused by sharp angles).

Figure 28. FE model of the original corner fitting with equivalent Von Misses stress values. The corner experiences the highest stress value of 1.51 x 10⁸ Pa. Note: the scale values are in Pascals.

Figure 29. FE model of the modified corner fitting with equivalent Von Misses stress values. The corner experiences the highest stress value of 3.2 x 10⁷ Pa. Note: the scale values are in Pascals.

The safety factors for the both fitting configurations were also investigated. It was predicted that the original and new fittings have a factor of safety of 2.3 and 7.8, respectively. The results are presented in Table 3.

Table 3. Maximum Von Misses Stresses and Corres ondin Safet Factors

Conclusion

Finite element analysis was performed on the original and new corner fittings to investigate stress concentrations and safety factors. Using the output forces generated from the overall cage frame analysis, both fitting styles were loaded on the diagonal pipe segment. The maximum stress was found near the diagonal - horizontal pipe joint. It was found that neither fitting would fail under the tested extreme load case. However, it was determined that the new, modified fittings have less stress under load and a higher safety factor than the original fittings.

ENGINEERING DRAWINGS AND SPECIFICATIONS

Detailed engineering drawings and specifications for the corner fittings and corner supports are shown below.

COPPER NETTING EVALUATION

OVERVIEW

The USB OCAT cage system was designed to use standard netting materials for the net chamber. Most standard netting materials, like nylon, have to be coated with an anti-fouling agent to reduce the possibility of marine biological growth. This bio-fouling reduces the water flow through the cage and also increases the drag of the system. One new technology that can eliminate this problem is copper alloy net chambers. The alloy resists marine organism growth, which makes it a solution to the bio-fouling dilemma. However, how these metallic nets affect cage systems is not fully understood.

A study was conducted by the University of New Hampshire Center for Ocean Engineering, in conjunction with the Mechanical Engineering Department, to better understand the drag characteristics of copper alloy net and to investigate the possible future use of this netting on the OCAT cage system (DeCew et al., 2007). To do this, the research was split into two areas of interest. First, a series of tow tests to determine the drag characteristics of copper alloy and galvanized steel net panels were performed. The drag forces for 1 -m² net panels were found for seven different meshes. The coefficient of drag for each net was also determined.

The other area of research focused on predicting the cage systems' dynamic and mooring load response in waves and currents with two types of copper alloy netting. Since the OCAT cage was originally designed to be used with a nylon net chamber, the cage was modified to support the heavier net, and any necessary adjustments to the net pen's hydrostatics were determined. The cage and mooring system were then modeled using a UNH developed program, Aqua-FE. This program can simulate waves and currents on structures in a marine environment, and the motion response and mooring line loads can be predicted. The cage system was tested under three current profiles, three wave regimes and a combination of both. A comparison was then made to an OCAT net pen incorporating a nylon net. The results were used to help determine whether this type of cage system is a candidate for copper alloy netting.

NET CHARACTERISTICS

Two types of copper alloy netting were analyzed for this study. Table 1 lists the net characteristics. For comparison, a typical nylon net used with the OCAT cage system is also presented. The two netting properties investigated bracket possible deployment scenarios of the cage system. Net type 1 is intended for a 1-2 year deployment, and smaller fish could be raised. Net type 2 would be utilized for a grow-out of mature fish for a longer term deployment (> 2 years).

METHODOLOGY

The purpose of this study was to predict the cage system's dynamic and mooring load response, with the copper alloy netting, in waves and currents, and to determine if the OCAT cage system is a good candidate for these net types. To do this, the hydrostatics of the unaltered cage (with nylon netting) was first determined. The same calculations were then performed with the two copper alloy nets installed on the system and compared to the unaltered cage. It can be expected that the increased weight of the copper netting will affect the hydrostatics and possibly the dynamic and mooring line load response of the net pen. However, the increased weight of net may significantly alter the center of gravity (CG) and center of buoyancy (CB) such that the system becomes unstable (even if no waves or current are applied). Therefore, to help insure that proper comparisons to the original cage type could be performed, effort was made to match the CG location and the total reserve buoyancy of the cage by modifying parameters such as the amount of ballast water and framework pipe size. Although the net pens may have an increased mass with the copper netting, the resulting systems will have similar metacentric heights and resting waterlines. This should allow for an easier comparison of how the metallic nets affect the cage and mooring system in waves and current. The cage modifications necessary to obtain these results are also described.

Once the hydrostatics of the two copper alloy netting systems were finalized, a numerical model was constructed. This model was subjected to three current velocities, three regular wave regimes, and a combination of waves and current. The load cases were selected for the following reasons: (1) for comparison to a previous study of an OCAT system with nylon netting; and (2) to determine the response of the system to conditions that a cage would experience in a marine environment. Table 2 lists the load cases.

The cage motion response in heave (vertical) and mooring line tensions were obtained from the simulations. The vertical motion was predicted from averaging the vertical displacement of two cage locations around the lower rim. In addition, the mooring line loads from critical components were also found. The chain near the anchor, and the bridle line loads were recorded from their associated elements in the model. To insure that the system reached steady state, the simulation was run for 800 seconds with a time step of 0.01 seconds.

HYDROSTATICS

The first step to determine if the OCAT cage system can support copper alloy netting is a hydrostatic analysis. The unaltered cage's (with nylon netting) mass, CG, CB and reserve buoyancy were first determined. The same calculations were then performed with the copper alloy net on the same cage framework. The results of the hydrostatic analysis are listed in Table 3. Note that the overall mass of the systems includes ballast water in the lower rim and side pipes consistent with the original design.

It can be seen that the addition of net type 1 would have little effect on the mass (increase of 48 kg) and center of gravity of the system. The added net resulted in a decreased reserve buoyancy of the cage. Net type 2, however, had noticeable effects on the cage's CG and reserve buoyancy. The net added approximately 1100 kg to the system's overall mass. In fact, the system would not float in its present design with this heavier netting.

To compensate for the extra mass and reduced buoyancy of both net types, modifications were made to the cage framework and ballast water in the pipes. For net type 1 (25 mm bar length), water was removed from the diagonal pipes so they were approximately 37% full (the original design had the pipes approximately 50% filled) to increase the reserve buoyancy and decrease the CG. The result was that the center of gravity was raised to within 1 cm of the original system. The cage with the second net type, however, needed to go through a series of design iterations before the hydrostatics were finalized. Changes considered included purging all the water from the pipes, increasing the amount of ballast chain to lower the center of gravity and increasing the HDPE pipe size. The result was a system in which the angled pipes were increased one pipe size and purged of all water. The amount of water in the lower rims was also decreased by approximately 10%. These modifications do not significantly change the construction, assembly and performance of the unit, thus were considered acceptable. The hydrostatics resulting from the cage modifications are listed in Table 4.

NUMERICAL MODELING

A numerical model representing the OCAT system was constructed using the Aqua-FE finite element program. Figure 1 shows the system with net type 1 (25 mm bar length) subjected to a 50 cm/s current and waves with a height and period of 1.63 m and 4.7 seconds, respectively. Figure 2 shows the two cage systems, superimposed for visual purposes, under a 150 cm/s current (no waves applied).

Figure 1. AquaFE results of a 25 mm bar length copper netting on the OCAT fish cage subjected to a 50 cm/s current and waves with a height and period of 1.63 m and 4.74 seconds, respectively.

Figure 2. Snapshot of both the 40 mm bar length cage as well as the 25 mm bar length cage subjected to a 150 cm/s current (3 knot).

The mean heave response of the cage and the maximum mooring line tensions predicted by the numerical model are listed in Tables 5 and 6. For comparison, the results for the unaltered cage system are listed in each table. The mean displacement is the vertical movement from the cage's static position where the top cage rim is located at the surface. The maximum load was found for three of the main mooring components: the anchor chain, the upper bridle line and the lower bridle line.

Table 5. Numerical model heave test results for the three net types.

DISCUSSION

In current, the OCAT cage system with the 25 mm bar length net chamber did not submerge as deeply as the 40 mm bar length net chamber. The maximum displacements of the net types 25-mm bar and 40-mm bar were -15.6 m and -17.3 meters, respectively. Although the cages are designed to submerge, the ballast chain on the system with net 40-mm bar net chamber did come in contact with the seafloor. If this system was to be deployed in areas with a higher water velocity, buoyancy may need to be added to the cage framework.

In waves, both systems responded similarly, remaining near the surface. Net type 2 (40-mm bar) did have a slightly lower heave response. In addition, the cage heave amplitudes of both systems were less than the wave amplitudes. This provides evidence that the cages are damped systems (not wave followers), allowing the waves to pass through the cage frame and net chamber. This can be attributed to the low reserve buoyancy and the mass (or inertia) of the heavier system, resisting the wave's oscillatory motion.

The mooring line tensions for 25-mm bar net were consistently less than that for the 40-mm bar net. One reason for this is the difference in solidities (25 mm bar length has a solidity of 7.4% compared to 17.4% for the 40 mm bar length). The maximum recorded tensions, occurring during load case 3, were 24.21 kN and 43.44 kN for the cages with 25-mm and 40-mm bar netting, respectively. In load case 7 (both waves and current), the cage systems have similar heave responses, however, the mooring line tensions for the 40-mm bar net were 61% larger. One interesting note is that the lower bridle tensions were larger than the upper bridle tensions. This is due to the length of the lines and the cage geometry. As the system sets back against the waves or current, the lower bridle lines absorb a larger percentage of the load. The unaltered cage mooring line tensions, however, fall between the two tested net types. One reason for this is that the twine net solidity is 10.2%, which is bracketed by the two copper alloy net solidities.

Recall that the same mooring components were used on all three systems. To insure that the mooring gear would be satisfactory, the safety factor was calculated for the cage system with the largest loads (40-mm bar net), assuming that the anchor does not drag. The anchor in aquaculture systems is sometimes intended as the "weak-link" in a mooring, designed to drag if the load is extreme, reducing the possibility of a component failure. The anchor chain has a minimum breaking strength of 266 kN (59,700 lbf) resulting in a safety factor of 6. The upper and lower bridle lines have safety factors of 10 and 5.5 respectively. These are all adequate. However, if the system may experience more extreme environmental conditions than tested in this study, the equipment may need to be upgraded.

The maximum anchor load for the simulations was approximately 44 kN (9891 lbf), where the anchor has a weight of 49 kN (11000 lbf). The holding power of deadweight anchors depends upon the shape of the deadweight and the bottom sediment. There is a chance the anchor could slide in small increments during the peak loads in load case 3. Therefore, the deadweight anchor may need to be resized once the bottom conditions are known.

Both copper alloy nets are viable options for the OCAT cage system. The 25-mm bar length netting requires no structural modifications to the original design, just a reduction of ballast water which can be easily performed. The system response in heave and the mooring line loads are less than that for the twine netting. The addition of copper alloy netting similar to the 25-mm bar net tested is feasible. The 40-mm bar length netting requires the cage framework to be altered (increasing the diagonal pipe size) and ballast water levels to be adjusted. The increase pipe size will require redesigning the original corner fittings. If these modifications can be incorporated into the cage framework, the heavier netting can still be utilized. The bar length and wire diameter of any net used in the field should be selected such that the net acts as a containment mechanism and not be oversized, thereby not allowing the fish to escape. In the mooring, the deadweight anchors holding power should also be increased.

MECHANICAL FEEDER OVERVIEW

The system is intended for use in the 100-m³ volume OCAT cage. Fish are presently fed floating feed by hand when the OCAT cage is at the surface. Feed is delivered through the top of the cage into a 4-m² enclosure that contains and prevents the feed from escaping the cage. This limited feeding area, coupled with the present hand delivery method, works well with pompano and other marine fish with a body shape similar to pompano. However, the system may not work well for species with a more streamlined body form. Feeding could potentially become space limited, particularly as fish size increases. An alternative system that can deliver feed into cage's interior volume was therefore desired.

In response to the ASA-IM request, UNH designed, fabricated and tested a portable feed delivery system. The design, associated modifications made during testing of the unit, and cost of the feed delivery system are described in the following sections. Assembly and calibration instructions, fabrication drawings, and a parts list are provided in the final report by UNH but are not included in this manual. They are available, on request, from the U.S. Soybean Export Council, 12125 Woodcrest Executive Drive Suite 140, St. Louis, MO 63141, USA.

DESIGN REQUIREMENTS

Performance requirements for the mechanized feeder were provided to UNH by USSEC (Cremer, 2006). These requirements specified a mid-cage mechanized feed delivery system that was simple and low cost in design, in keeping with the OCAT project objective of providing a low technology and low cost offshore ocean cage structure for small-scale fish farmers in Asia. Specific criteria included the following.

• Minimum feed bin capacity of 40 kg for fish feed pellets >3 mm in size

• Feed delivery tube that can be installed either through the 4-m feed enclosure or attached to one of the vertical cage frame pipes of the OCAT cage Feed delivery mechanism that will pulse the feed into the center of the OCAT cage to maximize fish access to the feed and minimize potential feed waste. The volume of feed delivered in each pulse should be adjustable.

Feed system capable of delivering feed to a water depth of 5 meters

Feed delivery system capable of delivering up to 120 kg of feed within 30 minutes

Feed delivery system must be a self-contained, portable unit with either a gasoline or diesel power source

The feed delivery system must be easily accommodated on a 5-m fiberglass skiff

The feed delivery unit must be capable of being constructed from off-the-shelf components and/or easily fabricated components.

DESIGN CONCEPT

The OCAT feeder design is based on the use of a venture system to create a low pressure area that pulls feed into flowing water from a pump (Figure 1). Only a standard water pump is needed in this approach because the feed is added to the water after the pump. This method allows for a much simpler design, but because some air can be sucked into the line, there is a possibility of building occasional back-pressure in the feed delivery hose (going to the cage). This method has been used to feed submerged cages at the UNH Open Ocean Aquaculture (OO A) site with success and little difficulty.

Figure 1. The feed venture system utilizes a standard water pump which moves water through a modified fitting before heading to the fish cage. The flowing water creates suction at the outlet of the internal pipe, helping draw the feed into the water.

To help ease the transportation and operational logistics of the feed delivery unit, the system was designed in individual units: water supply pump, water/feed interface unit and the dosing mechanism. The water pump unit consists of a standard pump, a water inlet hose, and water outlet hose. A ball valve can be added for pump priming, if necessary.

As the water enters the feed/water interface unit, it is directed through a set of 90 degree elbows and continues down a flexible hose to the cage. To allow for the feed to be mixed with the water, the first elbow is modified to have an inlet. A small pipe is threaded into the inlet and acts as the feed exit (Figure 2). As the water flows through the tee and around the interior pipe, it creates a low pressure area at the interior pipe exit. This helps draw the feed into the moving water, which is important when dealing with certain feed types and helps to avoid clogging.

I ll

Figure 2. The feed - water interface assembly. Water flows through a modified 90 degree elbow en route to the cage. This modified fitting generates a low pressure area and allows the feed to be added directly to the flowing water.

A knife gate valve and manual feed dispenser is located above the modified 90 degree elbow. The gate valve controls feed dosing into the water and controls back pressure that may build in the flexible hose during the feeding process. A temporary feed hopper (large funnel), is located directly above the valve and collects the feed (added by hand or automated mechanism) before it drops into the flowing water. The hopper separates the feed collection from the automated dosing mechanism (rotary or auger) for ease of construction, maintenance and portability.

The final unit consists of the mechanized feeding device. An auger was chosen in this design due to the required mass of feed needed to be delivered. Few, if any, dosing mechanisms dispense 120 kg of feed in 30 minutes using a reasonable amount of power that can be found on a small vessel. This unit was also designed as a stand alone component, such that manual feeding, without the automated auger, is an option.

OCAT FEEDER DESIGN

The first prototype, seen in the Figure 3, was constructed and tested in winter 2006. The system operated similarly to the concept described in the pervious section. A pump provided water to deliver the feed. The water passed through the feed/water interface unit. The manual and automated components of the unit were stacked above the feed/water interface frame. A manual collection hopper, with knife gate, rested directly above the venture system. Here, feed could be added by hand to the collection hopper and dispensed into the water via the knife gate valve. The automated portion of the feeder rested on top of the feed collection frame. This unit consisted of an off the shelf hopper and auger system. A control device would dispense the feed into the manual hopper before it was mixed into the water.

Figure 3. Assembly of the first prototype feeder unit. A manual feed collection hopper was located above the feed/water interface unit. The mechanized feeding unit was stacked on top. This resulted in the system having a high center of gravity, making it susceptible to tipping.

This system had a small footprint, ideal for a small vessel, and tested well, delivering the required feed rates. However, it possessed an operational difficulty for a small vessel. The unit had a high center of gravity, thus it was susceptible to tipping over. In addition, the unit height of approximately 6.25 ft (1.9 m) made adding feed into the hopper difficult. Therefore, the system was redesigned to reduce the overall height, center of gravity and increase its stability.

Design Modifications

In an effort to improve the operating logistics, the feeder was reworked in 2007. The mechanized dosing frame components were discarded for a floor based unit.

A free-standing hopper and auger system, as shown as an assembly in Figure 4, lowered the center of gravity of the system as well as kept the units portable. The feed/water interface components, including the water pump, were kept the same due to its success in the field. A 3-inch (7.6 cm) diameter vertical auger was selected to drive the feed to the temporary feed hopper. The auger is powered by a motor attached to the top of the auger and located within the PVC piping above the output fitting. To control the auger, a new motor control box was designed and fabricated. The control box, assembled in a water-tight housing, allowed for variable feed rates and was powered by a 12 volt battery.

Figure 4. The mechanized feed delivery components were redesigned to incorporate a free standing hopper and auger assembly. The original feed water interface remained the same due to its success in the field.

The hopper size was initially increased to a standard 55 gallon (208 liter) container. This allowed for the full 120 kg of feed to be held in the tank at once. The feed flowed from the hopper to the auger via PVC components: a 45 degree fitting, a section of pipe, and a 45 degree wye. This allowed for efficient use of space below the tank as well as a mechanism to utilize gravity to drive the feed into the auger. However, the PVC assembly required for the auger input, coupled with the large size of the hopper, resulted in a system height of 5.5 feet (1.68 m). This then made filling the hopper with 50 Ib (23 kg) bags of feed difficult. Therefore, the tank was modified to rotate 45 degrees (Figure 5). Though this made accessing the hopper lid simpler, there was concern regarding the connection below the hopper (a flexible connection was necessary which was prone to feed jams).

Figure 5. The hopper tilting mechanism, shown in the upright position (a), and the lowered position (b). This modification allowed for easy filling, the flexible bottom connections were susceptible to feed jams.

Since reliability of the equipment and operational logistics of the system were critical, the design was modified to utilize a smaller feed hopper. To allow for interchangeability of parts and assemblies, a tank of similar dimensions (diameter, cone angle) was selected, reducing the storage capacity to 30 gallons (114 liters). The height of the system was reduced by one foot (30 cm). This simplifies the hopper filling and requires hopper refilling during the feeding process. The assembly is shown in Figure 6.

Figure 6. Feed delivery system with smaller feed hopper and increased pipe size. The pipe at the outlet of the hopper was increased in diameter to reduce feed jams.

This system was then tested to insure adequate feed dispensing rates. Recall that 4 kg/min of feed was required for the design criteria. It was found that the sections below the hopper to the auger housing caused feed jams when tested with larger feed pellets (10 mm). Therefore these components were increased in diameter to reduce clogging and enhance transportation of the feed. Testing demonstrated that the system could now output more than 5.5 kg/min.

Testing the unit with different feed sizes revealed that feed would be stranded in the auger due to the vertical configuration, i.e. the auger would not empty itself. This could cause problems in the field, as the feed could spoil or clog the system in between feedings. Therefore the control system was modified to allow the motor to operate in forward and reverse directions. In addition, a feed outlet was added to the bottom of the auger base. A sliding door was cut into the supporting components that, when needed, could be opened to allow the auger to dispense the feed at the base of the system (Figure 7), eliminating the clogging potential between feeding events.

Figure 7. Reverse mechanism used to empty auger (transparent acrylic at bottom). A "door" was added to the bottom auger housing, allowing the unit to empty itself in-between feedings.

FINAL DESIGN

Upon completion of the modifications to the feed delivery system, each component was re-tested to insure that it worked properly. Figures 8-20 show the whole system and individual assemblies of each unit. The final design configuration is shown in Figure 21.

Figure 8. Feed delivery system with auger control box. The feed hopper stores the feed until needed. The auger then drives the feed up and outlets above the feed/water interface unit. Note: water pump is not shown.

Figure 9. A standard water pump is used to supply the water to transport the feed to the cage (viewed from above).

Figure 10. The feed water interface unit. This stand-alone unit contains the venturi that mixes the feed and water before it is pumped down to the cage.

Figure 11. The lower feed water interface components. Standard fittings were used to form the interface piping. Note: hardware not shown.

Figure 12: The upper portion of the feed water interface components. This portion allows for manual feeding by using the knife gate valve to control the amount of feed entering into the water. Note: hardware not shown.

Figure 13. The water/feed interface components prior to assembly in the framework.

Figure 14. The fully assembled feed collection frame and feed/water interface frame.

Figure 15. The hopper assembly. An off the shelf hopper is mounted in a frame to store the feed for the system.

Figure 16. The auger and housing assembly. The vertical auger is housed in standard PVC piping. It is attached to the hopper frame with supports. The system has two outlets: near the top of the system for the feed to be dispensed to the collection hopper and the bottom of the system for cleaning out the auger.

Figure 17. The auger assembly. The 3-inch auger is housed in PVC pipe. This pipe is attached to a wye fitting via an adaptor ring. The feed travels from the hopper, through the 45 degree fitting to the base of the auger. Note: Hardware and auger fitting are not shown.

Figure 18. The auger motor assembly. The motor is attached to the auger shaft via a coupling. It is then attached to the housing with a set of adapter fittings. Power is supplied to the motor through the PVC cap. Note: Hardware and auger fitting are not shown.

Figure 19. Mechanized dosing mechanism fully assembled. Clear PVC was used for testing purposes only.

Figure 20. The auger motor control box in a water-tight housing. Power is supplied with a 12 V batter. The voltage to the auger motor can be controlled with the variable control knob.

Figure 21 : The final design of the feed delivery system.

DESIGN EVALUATION

The feed delivery system underwent a series of tests to insure that the system was working properly. First, experiments were conducted to verify the proper feed delivery rates. The feed rates were verified on two different feed sizes: 6mm and 10 mm. It was found that the maximum output for the auger device exceeded minimum specifications, however, the feed rates did change between the two sized feeds. Therefore, it is important to calibrate the auger system with different sized feeds to insure proper feed delivery rates. The motor control was also subjected to a series of tests.

Once the feeding rates were verified, a test was performed to investigate the feed water interface unit susceptibility to back-pressure. The feed collection frame and water interface unit was set-up at the UNH Ocean Engineering tank. Water was pumped through the system (without feed) down to a depth of 20 ft (6 m). The knife gate was opened for various lengths of time without any problems.

A final test was performed to investigate the performance of the feed-water interface unit to combine the feed with the flowing water and send it down the delivery hose to a submerged cage. The feed delivery system was assembled, and brought aboard the R/V Meriel B to the UNH OOA site located 10 km from Portsmouth, New Hampshire in 52 meters of water. The prototype was then used to feed 50,000 cod, located in a submerged cage 10 meters below the surface, via a feeding tube length of 61 m. However, due to this long length of hose, and limited head pressure the purchased pump could provide, the vessels water pump was used. The manual feeder performed well in this test with no problems. In fact, a 20 kg bag of feed, placed directly in the temporary feed hopper, can be mixed into the water in less than one minute (Figure 22).

The automated portion of the unit was also tested. It was found that the improvements made during the design process solved any pre-existing conditions. The auger primed efficiently and feed jams did not occur with the larger pipe size. The auger emptying mechanism also worked well.

Figure 21. The unit was first tested by hand feeding, where the manual feed hopper was filled, and the knife gate valve opened in different increments (¹A opened, Vi opened, etc.) to obtain different rates of feeding.

COST ESTIMATE

The cost of the entire unit is estimated to be $4,420 and $2,120 for a system with and without the auger system, respectively (December 2007). Table 1 lists the cost of each section. The final cost does not include cost of personnel to assemble the unit.

Table 1 : The cost summary for the feeding system assuming an auger and controls were and were not incorporated into the design.

* All manufacturing/machining, welding rates assume a standard hourly rate of $60/hour.

The purchased parts used in this design were obtained from a number of easily accessed sources. It is possible that parts can be found less expensive from other vendors. Also, the cost listed here is for a single prototype. If multiple systems are constructed at the same time, the cost could drop significantly. The total cost does not include shipping NOTES

• The modified 90 degree fitting was constructed using a milling machine, with a 90 degree drill and hole saw. However, a CNC machine can be used for easy, accurate construction. A torch could also be used, if a jig is made.

• A series of two 45 degree elbows, placed before and after the modified 90 degree elbow, can be used to reduce back pressure.

• The base of the hopper system should be tied down to avoid tipping.

XII. REFERENCES

Celikkol, Barbaras, Judson DeCew, M. Robinson Swift, Igor Tsukrov, Ken Baldwin, Ashley Risso and Ryan Despins. 2007a. Numerical Modeling of the OCAT Cage and Mooring System. University of New Hampshire final report on USB Project 7512 to the United Soybean Board and U.S. Soybean Export Council, St. Louis, MO.

Celikkol, Barbaras, Judson DeCew, M. Robinson Swift, Igor Tsukrov, Ken Baldwin, Ashley Risso, Ryan Despins and James Wright. 2007b. Structural Modeling of the OCAT Cage. University of New Hampshire final report on USB Project 7513 to the United Soybean Board and U.S. Soybean Export Council, St. Louis, MO.

Celikkol, Barbaras, Judson DeCew, M. Robinson Swift, Igor Tsukrov, Ken Baldwin, Ashley Risso, Ryan Despins and Thomas Morganthau. 2007c. Physical Modeling of the OCAT Cage and Mooring System. University of New Hampshire final report on USB Project 7514 to the United Soybean Board and U.S. Soybean Export Council, St. Louis, MO.

Celikkol, Barbaras, Judson DeCew, M. Robinson Swift, Igor Tsukrov, Ken Baldwin, Ashley Risso, Ryan Despins and Jim Irish. 2007d. Construction and Deployment of an OCAT Cage in the Gulf of Maine. University of New Hampshire final report on USB Project 7515 to the United Soybean Board and U.S. Soybean Export Council, St. Louis, MO.

Cremer, Michael C. 2006. Pulse Feed Delivery System for USB OCAT Ocean Cage. Specifications provided to UNH by the U.S. Soybean Export Council, September 2006.

Cremer, Michael C, and H. R. Schmittou. 2003. Marketing Proposal: Ocean Cage Aquaculture Technology. Proposal submitted to the American Soybean Association and United Soybean Board, St. Louis, MO, 15 January 2003.

DeCew, Judson, Ryan Despins, M. Robinson Swift, Barbaras Celikkol, and Ken Baldwin. 2007. Preliminary Investigation of Utilizing Copper Net on a Small Scale Aquaculture Fish Cage. Open Ocean Aquaculture Engineering Project, University of New Hampshire, Durham, NH 03824. September 2007.

Goudey, Cliff. 2004. Cage and Mooring Specifications for the ASA 2 m x 4.5 m x 7 m Prototype Offshore Cage. Contract report prepared for the American Soybean Association, St. Louis, MO.

Schmittou, H. R. 2003. Cage Design and Construction. Design recommendations submitted to the aquaculture team, American Soybean Association, China, 10 November 2003.

Schmittou, Thomas. 2005. Engineering Report on the Cage and Mooring Specifications for the ASA 2 m x 4.5 m x 7 m Prototype Offshore Cage (OCAT Cage), 5 June 2005. Final report submitted to the American Soybean Association, St. Louis, MO. [00076] While various embodiments of the present invention have been described above, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of the present invention should not be limited by any of the above-described exemplary embodiments, but should be defined only in accordance with the following paragraphs and their equivalents. While this invention has been particularly described and illustrated with reference to preferred embodiments, it may be understood to those having ordinary skill in the art that changes in the above description or illustrations may be made with respect to formal details without departing from the spirit and scope of the invention.

Claims

Claims:

1. A fish cage, comprising: a frame comprising at least a first rigid tubular member; a port located on the first tubular member, the port adapted to seal a fluid within the first tubular member; and a non-collapsing containment net supported by the frame; wherein the buoyancy of the frame is adjustable by varying the amount of fluid located in the first rigid tubular member, the fish cage being coupled to an anchor at seabed, the coupling to the anchor defining a moving radius within which the fish cage is contained, wherein the buoyancy of the frame is adjusted to allow the fish cage to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

2. The fish cage of paragraph 1, further comprising a feed containment area including a plurality of splashboards and a feed containment net containing the feed, the feed containment area being mounted on top of the frame.

3. The fish cage of paragraph 1, further comprising a ballast attached to the frame, the ballast being adapted to offset at least a portion of the buoyancy of the frame.

4. The fish cage of paragraph 1, wherein the first rigid tubular member comprises at least one valve for allowing fluid in or out of the first rigid tubular member.

5. The fish cage of paragraph 1, wherein the first rigid tubular member comprises: a first valve arranged near an upper portion of the first rigid tubular member that allows the fluid to flow in or out of the first rigid tubular member; a hose coupled to the first valve on one end that stretches near a lower portion of the first rigid tubular member; and a second valve arranged near the upper portion of the first rigid tubular member that allows air to move in or out of the first rigid tubular member, wherein the amount of fluid located in the first rigid tubular member is varied by adding or removing fluid through the first valve.

6. The fish cage of paragraph 1, wherein a netting panel of the containment net includes a sealable gap, the gap being openable and connectable to a tunnel through which the fish are transferred to a second fish cage.

7. The fish cage of paragraph 1, wherein the frame is shaped as a truncated pyramid.

8. A single point mooring system comprising: an anchor arranged at seabed; a mooring chain coupled to the anchor at a first end; and a floating element including a first rigid tubular member coupled to a second end of the mooring chain, wherein a buoyancy of the floating element is adjustable by varying the amount of fluid located in the first rigid tubular member, the mooring chain defining a moving radius of the floating element, wherein the buoyancy of the floating element is adjusted to allow the floating element to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

9. The single point mooring system of paragraph 8, further comprising at least one bridle coupling the floating element to the second end of the mooring chain.

10. The single point mooring system of paragraph 9, further comprising a connector coupling the at least one bridle to the second end of the mooring chain.

11. The single point mooring system of paragraph 10, further comprising a buoy coupled to the connector, wherein buoy being adapted to offset at least the weight of the connector.

12. The single point mooring system of paragraph 8, further comprising a ballast coupled to the floating element to offset at least a portion of the buoyancy of the floating element.

13. The single point mooring system of paragraph 8, wherein an upper portion of the chain is lighter relative to a lower portion of the chain.

14. The single mooring system of paragraph 8, wherein the floating element is a fish cage including a frame shaped as a truncated pyramid, a containment net positioned inside the frame, and a feed containment area positioned on top of the frame, wherein the first rigid tubular member includes a sloping member of the frame.

15. The single mooring system of paragraph 8, further comprising at least one quick release buoy attachable to the floating element, the buoyancy of the floating element being adjusted such that the floating element submerges in the water substantially close to the water surface when the at least one quick release buoy is detached from the floating element, but float at least partially on water when the at least one quick release buoy is attached to the floating element.

16. The single mooring system of paragraph 8, wherein the first rigid tubular member comprises: a first valve arranged near an upper portion of the first rigid tubular member that allows the fluid to flow in or out of the first rigid tubular member; a hose coupled to the first valve on one end that stretches near a lower portion of the first rigid tubular member; a second valve arranged near the upper portion of the first rigid tubular member that allows air to move in or out of the first rigid tubular member, wherein the amount of fluid located in the first rigid tubular member is varied by adding or removing fluid through the first valve.

17. A mooring method, comprising: providing a fish cage including a frame comprising at least one rigid tubular frame member; and adding a liquid into the at least one rigid tubular frame member through a port, or removing a liquid from the at least one rigid tubular frame member through a port, thereby adjusting the buoyancy of the fish cage; wherein the fish cage is coupled to an anchor at seabed, the coupling to the anchor defining a moving radius within which the fish cage is contained, wherein the buoyancy of the frame is adjusted to allow the fish cage to automatically submerge in water with increasing wave or current action on water surface within the moving radius of the anchor.

18. The method of paragraph 17, further comprising: providing an anchor on seabed; providing a mooring chain coupled at a first end to the anchor; providing a connector coupled to a second end of the mooring chain; providing at least one bridle coupling the fish cage to the connector; and providing a buoy coupled to the connector to offset at least the weight of the connector.

19. The method of paragraph 18, further comprising providing at least one quick release buoy attachable to the fish cage, the buoyancy of the fish cage being adjusted such that the fish cage submerges in the water substantially close to the water surface when the at least one quick release buoy is detached from the floating element, but floats at least partially on the water when the at least one quick release buoy is attached to the fish cage.