DEVELOPMENT OF AN OYSTER CULTCH MATERIAL FOR THREE DIMENSIONAL OYSTER AQUACULTURE John Walker Fisher DEVELOPMENT OF AN OYSTER CULTCH MATERIAL FOR THREE DIMENSIONAL OYSTER AQUACULTURE by JOHN WALKER FISHER r J B.S. UNITED STATES NAVAL ACADEMY 1972 SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FCR THE DEGREE OF MASTER OF SCIENCE AT THE MASSACHUSETTS INSTITUTE OF TECHNOLOGY JUNE 1973 The, Library Naval Postgraduate School Monterey, California 93940 i DEVELOPMENT OF AN OYSTER CULTCH MATERIAL FOR THREE DIMENSIONAL OYSTER AQUACULTURE by JOHN WALKER FISHER Submitted to the Department of Ocean Engineering on May 14 in partial fulfillment of the requirements for the degree of Master of Science. Off bottom oyster culture is clearly superior to any bottom oed techniques, however it suf feres from the lack of a satisfactory cultch material. The recent practice by a large scale commercial Dyster aquaf arm of using hand punched scallop shells strung on a 1/8" nylon cord has resulted in estimated losses of 25-30% due to 3hell fracture and cord breakage. Drilling was investigated as a possible means of increasing useful shell processing and decreasing post-stringing losses. Based on laboratory experiments, a 1/8" steel drill bit operated at 1328 rpm was found to be more efficient than either 1/4" or 7/8" drill bits operated at any speed. A net processing time of 20 shells per ninute with a processing loss of 6.3% was possible using the 1/8" bit and drilling the shells one at a time. Group processing was considered, increasing successful output to 25.7 shells per minute with a processing loss of 18.9%. A polypropylene-crushed scallop shell composite was also con- sidered as a possible cultch and tests were made for abrasion and spat fall attraction. Abrasion was found to be of little importance, Spat fall was concluded to be a function of time after molding and surface texture. A commercially acceptable set was obtained in the laboratory by soaking the cultch in a 10% solution of fresh oyster shell liquor for three hours prior to 24 hour exposure. Thesis Supervisor: Dr. ira Dyer Title: Professor of Ocean Engineering 11 ACKNOWLEDGMENTS The author would like to express his sincere gratitude to r. Ira Dyer and Mr. Ray Pariser for cheir constant advice and ncouragement during the investigations. Their suggestions and uidance were invaluable in escaping many blind alleys. Special thanks go to Mr. Ed. Bia:.nard of Aquadynamics for posing he problem and giving so freely of his time in pursuing its solution, nd to Mr. Warren Landers of the National Marine Fisheries Service, ilford, Connecticut, who so generously volunteered his time and ex- ertise in conducting the laboratory setting tests. Without either f these men very little of any practical knowledge would have been ossible. Finally the author is deeply indebted to Ms . Catherine Kudla or her unending patience and encouragement in the preparation of the inal manuscript. To these, and the many others, regretably omitted, he author expresses his deepest gratitude. Ill TABLE OF CONTENTS )s tract. ... - . . .i ;knowledgements ii ible of Contents iii ltroduction 1 ff-Bottom Requirements 9 rilling 15 Dlypropylene Composite .17 Dmposite Cultch Evaluation 22 Lscussion 29 inclusion 32 ppendices I Wave Effects on Submerged Racks 34 II Cost Analysis 36 III Drilling Proceedure 40 IV Composite Cultch Proceedure 48 iterature Cited 58 ef erences 5© INTRODUCTION The commercial harvest of oysters in the United States las been falling steadily since the oarly 1900 's.l Siltafcion :aused by extensive dredging near many beds, ecological imbalance Leading to a rise in predators, coastal pollution, over harvest- ing or mismanagement of natural beds, and increased processing ^osts without the requisite price increase are all possible, and probable, causes for the downward trend. If oysters are to play a significant role in alleviating the anticipated world ood shortage of the future, better methods of oyster bed management and new techniques for improving population densities must be Derfected now. In the critical field of high yield aquaculture the United States is sadly lacking. Current yields (1968) of some of the worlds leaders in shellfish production are presented in Table [1] . Df the countries listed, only Japan produces oysters exclusively by off-bottom culture, though Spain and France are heavily in- volved in similar methods for cockles and mussels. The United States oyster industry obviously does not compare favorably with these countries. The dramatic difference is not attributed to climatic or environmental variation, but simply to the widespread application of the 17th century technology of off- bottom culture in Japan. Before discussing the commercial potential of an organism, 2 it is first advisable to understand its life cycle. Galtsoff 's monograph is a highly recommended source, although the American f TABLE 1 Aquaculture yield of cultivated stocks or natural popu- lations witii no fertilization or feeling. Units in fresh weight, shells excluded. COUNTRY U.S. U.S. France France Australia Australia ♦Japan (Inland Sea) Malaya Cockles France Mussels Philippines Mussels ♦Spain Mussels SPECIES Oysters (National Avg . ) Oysters (Best Yield) Flat Oyster (National Avg.) Portugese Oyster (National Avg . ) Oysters (National Avg . ) Oysters (Best Yield) Oysters kg/ha/yr t/acre/yr 9 0.004 5,000 2.00 400 0.16 935 0.37 150 0.06 540 2.20 58,000 23.30 12,500 5.00 2,500 1.00 125,000 50.00 300,000 120.00 Source: The Status and Potential of Aquaculture. Volume I. Particularly Invertebrate and Algae Culture. John H. Ryther and John E. Bardach. Calculations based on an area 25% covered by rafts. jyster, Crassostrea virginica Gemlin is a well studied, highly locumented creature. The American oyster reaches sexual maturity at 10-12 months # md may spawn any time conditions are favorable after that point. te ambient temperatures begin to rise above 20°c/ gonads are developed and egg and sperm released into the water, independent }f any mate. The eggs are fertilized outside the shell in Dpen water. These will usually hatch within 48-96 hours, pro- iucing larvae which are free swimming for the first 20-25 days. \t this point they metamorphose and set, attaching permanently to a handy solid object. The spat, as they are then called, lead a sessile life from this point on. Growth to marketable size iepends entirely upon temperatures of the surrounding waters, though it may be stunted by nutrient unavailability. In Massachusetts, this growth period averages 4-5 years for wild, oottom grown oysters. Shaw3 charted the growth of oysters in Massachusetts as a function of both time and temperature. (Fig. 1) He found that growth was directly dependent upon ambient water temperature and that virtually no shell growth occured when the temperature dropped below 10 °c. The oysters were growing only 4 months a year, thus requiring 3 times as long to reach market size as similar oysters grown in tropic waters. The justification for basing a large scale oyster farm in the cold Northern waters of Massachusetts is the higher quality oyster produced. During the winter months, when shell growth is 30 __ 20 - 10 _ Lgure l — Monthly increase in height and length of oysters as percentage of total growth during 1957. Temperature curve based on records taken at time of measurement. r retarded, the oyster adds fat or meat, improving the flavor considerably. This difference is quite noticable between the winter harvest oysters and oysters harvested any time of the year from warmer tropical waters . Bottom bed management in the United States usually in- volves little more than planting young "seed" oysters, killing predators and harvesting the crop, although some intensive manage- ment programs do include thinning the beds to reduce crowdinc and careful control of silt build-ups. The seed oyster, are small oysters generally grown in a separate operation designed specifically to produce only immature oysters and their pro- duction is an industry in itself. Once purchased, they are scattered over a prepared bed of cleaned oyster shells, to build up the bed. The seed attach to the shells, are allowed one or two years growth, depending on area, and are harvested regularly by several methods. Oyster mortality is quite high for bottom culture, running as high as 90% in some areas. Oyster drills and starfish can decimate a bed in short order, and they must be carefully mon- itored and controlled. Siltation can easily suffocate large areas of oysters, and is particularly dangerous to young oysters. High pressure water jets or suction jets are ofter used to control this problem, as is frequent transplantation. Off bottom oyster culture probably began with the Japanese, who started collecting spat on upright bamboo stakes sometime in the late 17th century. While this method is still actively used, most Japanese oysters are grown in raft culture. Wooden or bamboo rafts (50 ' X 60') b yed by hollow drums are used to support 500-600 wires, each 30-40 feet long and strung with * oyster shells on 7 inch centers. The wires, or rens as they are called once they are strung, are used to collect spat during spawning and then attached to the raft which is anchored in protected waters for the one year required to bring the oysters to market size. The rafts are generally quite crudely constructed and have a lifespan of only five years." Several United States growers have experimented with raft culture to gain the advantage of a high growth density, low mortality rate and ease of harvest. The results have been some- thing less than overwhelming. In an Alabama study, ' floating rafts were constructed of lumber, b yed with styrofoam, and six foot long scallop shell rens were attached. Raft damage was slight in the protected bays of Alabama, growth was rapid, mortality was lower and the costs were prohibative. To grow $162 worth of oysters (18 barrels @ $9/barrel) requires approximately $347 in labor and materials. This includes a very significant cost of $.75/ren or $147 for stringing the 196 rens required to harvest 18 barrels of oysters. A Massachusetts study in 1962^ experienced considerable problems with raft-environment interaction, with particularly detrimental effects on the rafts. Raft damage and loss of rens due to storm conditions were significant factors in the loss of much of their raft-borne culture. Even so, 3 2 bushels of oysters (worth $368 @ $11. 50/bushel) were produced in two years, at a cost of $248. The conclusions of the investigating team were generally favorable toward commercial raft culture, however, primarily as a means of exploiting the states1 seed oyster potential. A similar problem plagued an Oxford, Maryland study in 1970.9 During the winter storms, most of the 1/8" nylon string being used in the rens were frayed by constant movement of the shells, resulting in the loss of most of the rens. The jaggered edges of the punched holes in the shell were at fault. Damage to the rafts proper accounted for the remainder of the lost rens . In a Georgia feasibility study, 10 shell erosion and growth retardation were found to be major drawbacks to raft culture. A high silt concentration and water current velocity were concluded to be the previous cause. These problems could be corrected by flow deflectors and obstructions, though this would increase raft cost. As it was, this study concluded that, due to high labor costs and low dockside price of oysters, raft culture was not economically feasible in Georgia. In general, raft culture was not feasible in the South due to low prices and high costs, and in the North because o.' bad weather and cold waters. One factor not considered in either region, but . of great importance in the United States, is the availability of large expanses of surface water suitable for raft culture of oysters. The high demand for coastal waters for recreational sports places a high premium of those lands best suited for oyster raft culture. A second Japanese method of culture is the long line system. are moored 10-12 feet apart and connected by long, tarred rice ropes on which are hung rens spaced 3 feet apart. The primary advantages of this system are the low cost of con- struction and durability in rough waters. This method has been quite successful in Japan, particularly in those waters where raft culture ir. unfeasible due to high wind or wave conditions. It has not fared quite so well in the United States^ however.- In the 1962 Georgia studyll previously described, a long line experiment was also initiated for a feasibility study. After four months of constant immersion, the rens had become hopelessly tangled. The primary cause seemed to be a dramatic increase in weight ascribed to rapid growth of fouling organisms. This brought the buoys closer together and allowed the rens to interact with each other. Efforts to increase the interval distance between rens and to anchor the buoys , more firmly were futile and the experiment was abandoned. OFF-BOTTOM REQUIREMENTS In light of the previous discussion, three dimensional oyster culture in the Northern United States imposes two additional requirements on this system, peculiar to the area: (1) Due to the lack of free coastal surface waters, the farm must utilize either very little area t or leave the surface free for other use. (2) Special attention must be paid to environmental effects on the framework supporting the rens . Severe weather, high wind and wave conditions and winter ice must all be anticipated and designed for. One possible solution is to compromise with both surface and bottom culture methods and place negatively buoyant racks on bottom. The advantages are clear: » (1) By placing a six foot high rack in 15-20 feet of water, the upper water column is left free for recreational use while still providing a three dimensional environment in the lower strata. The result is increased population density, higher growth rates, lower mortality and easier, predictable modular har- vesting, with free surface loss held to a minimum. (2) B" removing the buoyancy requirement, a more stable configuration can be constructed of sturdier materials , such as steel or aluminum. The result is sturdier, more durable racks, more resistant to the marine environment. (3) By placing them on the bottom, the racks are pro- f 10 ected from surface waves and currents. The demonstration of this sing linear wave theory is included in Appendix I. Such a jystem of steel racks, placed on the bottom are c.irrent- y in commercial use a'ti Aquadynamics" , a Marion, Massachusetts aquafarm. reliminary results from the initial harvest indicate a requirement or a higher strength stringing material and an improved cultch. In 11 other respects the technique of submerged racks has proven to be n excellent one. The Aquadynamics farming system employes steel racks measuring l'X 7' X 6* to suspend approximately 200 rens off the bottom. Each an is 5.0' long and contains 20 hand punched scallop shells, strung n a 1/8" braided nylon cord, separated by 3" spacers made of Poly- inyl cloride. The rens are strung by hand and suspended in Ocean Dnd to collect a set prior to being suspended on the racks . At the resent time over 900 racks are sitting on the bottom in Buzzards Bay epresenting an estimated 40,000 bushels of oysters. After the three year period required to bring the oysters to arket size, the racks are winched aboard a special catamaran and car- ied back to a warehouse for picking, sorting and packing. The major- ty of Aquadynamics1 market is for half shell oysters, hence no shuck- ng is required. A number of minor problems have been e countered, primarily in he post-growing phases of operation. A better catamaran design is eeded to increase carrying capacity in stability. An automated eans of cleaning the shell exteriors, possibly by wet tumbling, would peed packing and reduce handling costs. 11 The major problem, though, has been with the durability o^ the ens. Cultch losses due to string failure and shell fracture are stimated at 25-30%. While these losses represent only a contain- ent loss and not an absolute organism loss (the oysters are still live on the bottom) , the higher mortality rate of bottom oysters rid the need for a second harvest do represent an important cost ncrease. At the present time no plans have been made to recover riese oysters, hence they represent a net loss of 25-30%. The object of this project was to develop an inexpensive nateri- 1 which would reduce cultch related losses and still fulfill all eripheral requirements of a successful cultch. The cost and price igures quoted are taken from the production records of Aquadynamics f rid thus represent true data from a large scale commercial operation, nly in this way could realistic cost limitations be applied. To be successful, a cultch material must be capable of attracting ommercially acceptable set, usually around 1-2 spat per square inch, ost of the work done on cultch development has centered on this spect as a yardstick against performance. The requirements for good cultch range far beyond the simple one of attracting an cceptable set, however. In addition, in a subjective order of tnportance: BIOLOGICAL RESISTANCE: The cultch must be relatively resistant to algae and other marine fouling organisms that may smother or crowd out the young spat, or must be capable of securing a quick set, without requiring long term immersion to season the material. FIGURE 2 ...._. — — , — , — _ , — _., . 12 i . g f tr - £ t u \ \ \ i .^fe£j ii rttrtS ffi Hand punched scallop shell ren currently being used 13 DURABILITY: Inertness to sea water and resistance to abrasion is very important. Since the rens will be immersed for taree years without maintenance and subjected to continual wave action, durability is paramount. STRENGTH: Tensile strength must be sufficient to support the weight of the matured oysters. Many otherwise acceptable mate- rials are so severly stressed during processing that they cannot be used. A significant portion of the 25-30% loss figure for Aquadynamics can be attributed to shell failure. COST: The original 1970 cost per ren was on the order of $.35 with a near future projected increase to $.40 due to additional costs of material acquisition. This has since been revised up- ward to its current estimated value of $.48. A new cultch ma- terial with a 0% loss rate could cost as much as $1.83 per ren without any loss in net income. (See Appendix II for cost breakdown and analysis.) A maximum cost of $.80 per ren was chosen as an upper design limit to allow application of the cultch to areas where the price of oysters is considerably lower. TOXICITY: Oysters are extremely efficient at concentrating trace pollutants, particularly harmful ones. Any toxins in the cultch that can be leached out by the sea water could easily be concentrated in the oyster meat, or kill the oyster. REUSABILITY: With the current emphasis on ecology it behooves an investigator to consider the possibility of recycling the cultch or of employing a recycled material in its manufacture. There is also an obvious cost reduction advantage in using such materials . 14 WEIGHT: Weight is of importance primarily in the setting stage handling of the rens . The rens are gathered in bundles of 20 and placed by hand in Ocean Pond to secure the set. They must be held head high to prevent them from dragging on the ground, hence an arbitrarily established limit of 2 lbs. per ren or 40 lbs. per bundle was imposed. The bundle size could be reduced if necessary, though this would probably reduce the set densz.ty. 15 DRILLING The search for the ideal cultch has generally been centered round the necessity of collecting a heavy enough set to bring a ommercial crop to harvest. For the present, scallop and oyster hells have been the most successful in attracting young larvae, nd thus have been the most widely used. Their use, however, re- uires some means of suspending them, which in turn effectively re- uires a means of manufacturing a hole. The most popular means is y simply punching a hole with a spik€*d hammer. While its inher- nt simplicity makes this method attractive, it also results in igh shell losses due to immediate fractures Cestimated 40-50%) . [ore importantly, many shells are severly stressed and weakened y the process but do not actually fail until heavily loaded with ysters. The rough, uneven hole also tends to severly chafe the lylon string, causing further losses over the 3 year span of growth, lorabined, these losses account for the 25-30% loss figure previous- .y quoted. As a possible solution, an attempt was made to drill the re- hired hole through the shell. It was felt that a cleaner, neater lole would reduce stress fracture and reduce chafing losses. Three Irills were tried, each from both sides of the shell. The raw data ind proceedure used are found in Appendix III. The following conclusions were drawn based on this data: (1) The 1/8" diameter drill was far superior to either the 7/8" or 1/4" drills and represented a feasible means of manufacturing a hole. 16 (2) There was no significant correlation between shell thickness as measured and ability to withstand drill processing. (3) Pre-sorting the shells is a desirable step in the processing chain of events to prevent wasteful processing of useless shells. (4) 1/8" group drilling has the potential of being an economical- ly viable means of lowering current shell processing costs, and of providing a stronger, more reliable cultch. r 17 POLYPROPYLENE COMPOSITE Alternative means of incorporating scallop shells into a cultch 2re also considered. The process envisioned entailed grinding the hell into an acceptable sized grit and using a binder to reform it n the desired shape. The natural choice for a binder was a plastic f some sort. There are several reasons for this choice. Plaf.tic s cheap, readily available, lightweight, quite durable in sea water, on-toxic, has a high tensity strength and an advanced state-of-the- rt manufacturing technology. The major drawback is its apparent liability to attract larvae, though there is conflicting evidence in this point. The North Carolina Department of Conservation and Development :onducted one of the few in depth field studies of both cultch laterials and techniques in 1967.12 They concluded that large [10" X 10") asbestos plates spaced 6" apart were superior to any combination of size and spacing using polypropylene either coated fith Inertol ( a tarlike substance) or with imbedded sandspots . fhey were also superior to smaller asbestos plates of all spacings . The sandspot-embedded polypropylene was an acceptable material for raising seed oysters, the experimentors found, but that spontaneous detachment after a few months rendered it unacceptable for raising oysters to maturity. The plastics experienced an initailly retarded set, but after being immersed for several weeks, a satisfactory set was obtained. Loosanoff1^ also found polypropylene to be acceptable to spat, but difficult for maturing oysters to bond to. Again, spontaneous 18 ttachment was a serious problem. Both of "-hese findings were in conflict with a 1962 study by he United States Marine Fisheries Service.14 In laboratory electivity tests, various cultches were tested in single competi- ion against scoured oyster shells for collecting spat. Their esults, presented in Table (2 J. show clearly that natural shell is tuch preferred to any synthetic material and that plastics Cpoly- ityrene in this case) are not satisfactory for collecting spat. ?o confirm the laboratory studies, field tests were carried out in Jie harbor waters of Milford, Connecticut. Several wire mesh sacks, jach containing one type of cultch and representing several dif- erent materials-, were spread randomly over the harbor during spawn- ing season. Thus, the larvae were free to choose among any of the materials. The chart in Figure (3) demonstrates once again the desirability of natural shell, and the unsuitability of plastic, as a setting medium. The combination of a rough scallop shell surface embedded in strong, durable polypropylene substrate was felt to be an ideal composite. The texture of the shell surface would be able to attract and hold the spat as they matured while the plastic main- tained the cultch integrity without fracture problem. To test the feasibility of constructing cultch in this manner, laboratory test specimens were developed and tested. Several variable parameters were investigated and their relative importance evaluated. Various concentrations of scallop shell filler were mixed with 19 Table 2 Number of oyster spat on artificial collectors and oyster shells obtained under similar conditions in laboratory tests Common name Average number of spat per ,6 sq. in. Oyster shells Artificial cultch -phenolic resin -vinyl plastison -silicone rubber -styrene/butadiene^ -styrene/butadiene^ -asphalt varnish -vinyl-latex -epoxy resin -polystyrene -acrylic resin -air-drying vinyl resin -natural rubber -polyethlene -polyester resin -tire stock -cement board-^ -cement board4 -insulating cement -neoprene -rubber cement abundant 35 48 131 85 16 284 171 89 8.7 86 125 103 246 187 487 63 244 114 0 0 0 1 1 1 1 1 1 1 1.7 2 2.5 2.3 13 37 49 16 75 56 ■■•Dow Chemical Co. 'Armstrong Rubber Co. Asbestoboard, Johns-Manville 4Flexboard, Johns-Manville 20 0. SHELL PERLITE1 CEMENT BOARD ABBCOBOND2 NEOPRENE USED TIRES TIRE STOCK NAT. RUBBER ASPH. PAINT POLYSTYRENE WITH 0. SHELL y////////////////y///////////zzm?, ■///////// ////////////////////M\ y///////////////////////m^^^ mi < ,- // 7777, 77, 4 6 SPAT PER 6 SQ. IN. 10 Figure 3. Number of oyster spat obtained on different materials under similar conditions in Milford Harbor, Connecticut. Iperlite-insulating concrete made from cement and peruxalite, a volcanic glass. 2Abbcobond-black rubber cement. 21 to types of polypropylene and hot compression molded using half- lell molds. Two methods of embedding the large shell grit into the jltch surface were also tested. The first, the so-called 1-step sthod, involved laying down a layer of grit in the mold prior to During in the polypropylene mixture. The second, the 2-step athod, required pre-molding of a polypropylene-filler disk and cheating to embed the grit. Detailed proceedures and raw data are resented in Appendix IV. 22 COMPOSITE CULTCH EVALUATION The durability of the composite surface was considered to be significant importance, primarily in preventing the maturing sters from spontaneously detaching or being jarred loose from e cultch. Abrasion caused by constant movement in sea water was mulated in the laboratory by a temperature bath shaker. Six rge disks molded from 5" diameter molds and evenly representing th methods of surface embedment were immersed in unfiltered sea ter at 22 °c and fixed to the shaker bed. The two different meth- s of surface attachment were placed in separate tubs in the same aker . The shaker bed described an approximately circular orbital th of 6" circumference and was operated at a frequency of 6Hz. nee this frequency of the shaker was well above the natural fre- tency of the water, the travel of the cultch represented a true locity through the water of 3r per second. The tubs were drained after 2 4 hours and filtered through dO filter paper with the aid of a vacuum pump to retrieve any iterial loosened from cultch surface. An equal volume of sea water is filtered through another filter paper to obtain a background >lids figure. The papers were dried and weighed, while the iltchs were returned to the shaker with a new volume of sea water >r another 24 hours. The water was again drained and the dried 23 ilter papers weighed. The difference between the filtered cultch water and the plain ;ea water w^s taken to be the weight of material lost from fie disk ;urf aces . For the first measurement these averaged 2.3 grams per lisk for the two-step group or .5 grams per disk for the one- step rroup, representing losses of approximately 2% and .4% respectively. he second measurement showed no appreciable surface grit in either ub. Based on these results, there appears to be little concern over lisintegrating surface texture. The velocity in the shaker vvas much ligher than can be reasonably expected in the interior of the racks, /here the oysters act as a break water. Mechanical abrasion caused )y spacer-disk interaction was not considered, since no spat would survive if caught between the two and it was not considered likely that either one would totally degrade the other. Although it was impossible to obtain a field set, a laboratory experiment was carried out with the help and cooperation of Mr. Warren Landers of the National Marine Fisheries Service. The six 5" disks ised in the abrasion study, a number of scoured scallop shells and two freshly molded (2 days old) 3.5" disks were placed in three trays according to Fig. 4 . Several thousand 3-week old larvae were added to the trays for 4 or 6 hour exposure. Three questions were considered: 24 M U) w •H in ■ in m i I CN iH / \ & / \ « a / \ +■ 0) / \ w +j 1 W \ /H I r ")" oV J V / Q \ / +* V y «« i CO s En o -o in rtf .£ revious tests were soaked in a 10% solution of FOSL for a period 26 TABLE 3 A/21113 4/29/13 4-hr. immersion 6-hr. immersion Pan 1 Large 1 step Spat ■5 Spat 0 Large 2 step Large 1 step 8 2 . 2 0 15 2 2 Total area: 5 8. 8 in Spat Density: .25 /in Total area: 5 8. 8 in2 Spat Density: . 03/in Pan 2 < Large 1 step Large 2 step Small 2 step 10 11 2 1 4 0 23 Total area: 48. 8 in^ Spat Density:. . 47/in2 Total area: 48. 8 in Spat Density: . l/in^ Pan 3 Large 2 step Small 2 step 3 3 0 10 0 10 Total area: 29. 9 in2 Spat Density: . 1/in Total area: 29. 9 in . Spat Density: . 35 /in Scallop Shells (1) (2) (3) (4) (5) (6) 35 29 3 174 135 78 7 4 52 22 139 70 454 294 Total area: 42. 6 in Spat Density: 10. 7 /in' Total area: 42. 6 in2 Spat Density: 6. 9 /in2 27 of 3 hours. The porosity of the cultch allowed penetration of the solution, enhancing retainment. of the FOSL* after withdrawal from the soakinc bins. The treated disks were placed in a tray wj_th several thousand larvae from the same group as was used in the pre- /ious setting experiments. Three disks exposed only to sea water rere similarly exposed as a control group in a separate tray. The comparative results after 24 hours exposure were quite dramatic, as Ls evident .^rom Table ( 4 ) . The treated disks were able to obtain a commercially acceptable set without any difficulty, while the con- trol disks secured approximately the same set density as before. / TABLE 4 28 Total # Spat Avg. Avg . previous 2 tests Control group (2 2-step, 1 1-step 5 top .085 large disks) 3 bottom .051 .14 (1) Treated group #1 small 25 top 2.6 >25 bottom >2.6 #2 small 10 50 top bottom 1.1 5.2 Large 2-step 50 20 top bottom 2.5 1.02 Totals Total area >180 77.1 3 .1 .33 2.3 spat per inch .19 NOTES (1) There was no appreciable bottom set in the first two tests, and no count was made. The figures thus reflect only the set on the top of the disks. The set in the third test was split approximately 50-50, hence, the bottom set could not be ignored. The probable cause for the discrepency is the different length of exposure periods. 29 DISCUSSION Based on the results of tables ( 3 ) and ( 4 ) and experience in he laboratory, several observations are possible. » (1) The cultches made by the 2-step technique appeared to be more successful than the 1-sfcep cultch in attracting spat. They also provided a consistently rougher surface, dis- couraging spontaneous detachment of maturing oysters. (2) While the 2-step cultches suffered more loss in the abrasion test, the absolute quantities involved could be considered negligible. Abrasion by sea water was not a major problem in the laboratory . (3) A 50% filler concentration is the maximum desirable dilution if crushed scallop shell is to be used as the filler. It is recommended that a lighter, cheaper more readily available material be employed in this capacity. (4) The flake polypropylene with 30% glass fibre was superior in every way to the pellet form, at least as far as the methods utilized for molding were concerned. Designed as a special homopolymer modified for chemical coupling with glass, it proved to form a superior bond with calcium carbo- nate as well. (5) To prevent ultra violet ray damage from prolonged exposure to the sun, a pigment of some form should be added. The coloring of potash may or may not be sufficient to block appreciable UV damage. A one year constant exposure limit should be used as the time factor for design and test work. 30 Aquadynamics requires on the order of 10 ^ disks per year to ointain the.lr present operation. Any expansion would entail even reater quantities. The laboratory techniques used to manufacture he test disk:s were deemed unsatisfactory to meet this demand, even J.f carried out on a large enough scale. (1) Inconsistant quality of the disks produced was a serious drawback. This may have been more of a problem with equip- ment and execution of proceedure than with the technique it- self, however there was an unacceptably wide disparity of quality between sequential disks produced by the same method. (2) The time involved in individually molding each disk and the expense of producing the molds for the quantity required, would be prohibitive. (3) The processing of molding-cooling-reheating is a wasteful and time consuming one, but necessary to remove the disk and embed the crushed shell using the 2-step method. Re- expansion of heated disks where the pressure is released is also damaging to quality control efforts. (4) The wide difference in coefficients of thermal expansion be- tween polypropylene and crushed scallop shell warped the thinner disks. The crush shell formed a tightly compacted surface layer on one side of the disk, while it was hot. As the cultch cooled, the plastic, with its higher coefficient of expansion, contracted much more than the compacted shell surface. The result was a convex shaped disk. While not considered to be a major drawback, such non-uniform effects may be an obstacle to automated stringing. 31 A more feasible means of mass producing the disks would require oiling out large sheets of polypropylene which are .05" thick. While till warm, these could be laid over a bed of heated shell grit and ressure applied by a large cookie cutter. . This would embed the hell in the disk and form the disks including the hole, simultaneously he excess plastic could then be returned to the melting pot for e-rolling into sheets. / 32 CONCLUSION There are several good reasons for choosing a polypropylene oraposite cl.1 ten over natural shell. (1) Polypropylene is far less likely to fracture or to abrade the stringing material. Post setting losses of cultch can therefore be expected to drop considerably. The current program of strengthening the stringing material attacks only one part of the problem, increases the cost of the rens and has not been field tested. (2) Manufacturing the cultch provides uniformity of disks, thus enhancing the opportunity for automated stringing. By eliminating time consuming hand labor, it is expected that costs can be reduced significantly. • (3) The entire quantity of shell paid for and transported to the stringing site is utilized. There is no loss due to pre- processing breakage or additional transportation costs for disposal of the unsuccessfully processed shells. (4) Coarse control of the spat fall may be possible through reg- ulations of the pre-soaking mixture of FOSL. This would provide an adjustable prarmeter to prevent over crowding. Should polypropylene be undesirable for any reason, such as prohibitive manufacturing costs, the utility of natural scallop shell :an be increased several ways. (1) Drilling the hole with a high speed 1/8" drill will increase the processing rate while reducing processing losses. (2) Pre-sorting the shells will eliminate unnecessary and waste- ful processing of broken shells. 33 (3) Post processing testing will help spot non visible failures and reduce post-setting losses. This can be done by the stringer without any appreciable time loss. (4) Vinyl-coated wire should be considered as a possible string- ing material to reduce losses from breaking strings. 34 APPENDIX I WAVE EFFECTS ON SUBMERGED RACKS According to linear wave theory, the velocity potential of a progressive water wave may be expressed as: \ ' to /. .1 u L 00 CosU fcU The horizontal particle velocity, u, may be written as or, since C: — ^ AL ^ kU-v) ^c**— *o **" r | *o) - Z.ol £+/s*c 36 APPENDIX II COST ANALYSIS callop Shell Ren Initial costs of the ren was approximately $.35, broken down as ollows: Materials string-1/8" braided nylon cord $.08 spacers-l/2"PVC tubing $.05 Labor punch hole & stringing (This is $.15 piecework. cost of punching the hole is estimated at ^$.02.) tying off the ends $.03 cut spacers $.01 transportation of shell and strings $.04 $.35 The sudden demand for scallop shell in the Buzzards Bay area had e effect of elevating it from the role of waste material to economic esource, resulting in a rise in acquisition .and transportation costs, he cost of vhe string also rose, primarily as a result of switching o 3/16" cord in an effort to reduce string loss. As of lMay 1973, stringing costs stood as follows: Materials shell $.10 h / 37 string $.10 spacers $.07 Labor stringing & tying (piecework) $.15 cut spacers $.01 transportation $.05 $.48 In addition to the stringing costs, a royalty fee of $.20 per ren is charged by Ocean Pond to obtain the set. Thus, the current /alue of a ren on the rack is approximately $.68. Yield estimates from the first harvest at Aquadynamics indicate an average yield of nearly .25 bushel per ren, at a current market yalue of $22 per bushel. From these figures a crude real-cost figure :an be calculated for the cultch. Value per ren = (.25) bushel per ren X $22 per bushel = $5.50 Loss per ren = (.25) (5.50) = $1.35 Cost per ren = $0.68 Net profit per ren = $3.47 Total cost per ren = $2.03 Subtract royalty fee = -$0.2G Total manufacturing cost per ren = $1.83 Any reduction in this figure represents a net profit increase, 2nd is most desirable. For the purpose of this paper, a ceiling of 38 $1.00 per ren was imposed. polypropylene Composite Considering assembly costs to be fixed, as a first approximation, maximum production sosts can be estimated. Approximately .025 lb. per disk polypropylene was required for the 3.5" cultch. This represents a cost of .5 lb. per ren or,, at a current bulk rate of $.16 per pound, a cost of $.0 8 per ren. Potash, if used as the filler can be expected to cost less .than $.02 per string, mostly for transportation. The cost of the unpros- sed scallop shell is difficult to estimate. Most purchases are made in units of dollars per pile, price being determined by non-standard- ized haggling. As an estimate, $.04 per ren will be used. Material costs, then, add to: Maximum allowable cost $1.00 Labor cost .21 Setting royalty .20 Spacers .07 Polypropylene .08 Filler .02 Shell .04 Maximum production cost $0 .30 This figure could probably be reduced at least two ways . Auto- mated stringing at the manufacturing plant would reduce the labor costs considerably; and the use of a vinyl coated wire as the stringing material instead of 3/16" nylon cord could help reduce costs, while boosting reliability. The major cost increase will be additional 39 treatment costs to soak the disks in FOSL prior to securing the set. / 40 APPENDIX III Drilling success was considered a function of four parameters: shell thickr.ess, drill diameter, drill speed and side of entry. In the initial test, a 1/4" steel drill bit was run at low speed (476 rpm) on a flat table drill press. Both convex and con- cave entries were attempted and recorded. Shell thickness and suc- cess were tabulated. The shells were presorted to remove any shells already fractured. Af terprocessing, the shells were tested b^ the application of a light bending moment. (1-2 in-los) as a check against non visible failures. The shell thickness was measured with a 1/4" diameter micrometer. This method was not totally satisfactory since it did not, in fact, measure shell thickness, but rather ridge depth as shown in Fig. ( 5 ) . The speed at which shells could be processed was considered an important factor in determining the economic feasibility of drilling. The number of shells successfully drilled over a two minute time span was used as a rough figure. It should be pointed out how very rough this figure really is. The experiment was carried out by an interest- ed investigator concentrating on his task over a relatively short time in the artificial conditions of a machine shop laboratory. On the other hand, the investigator did not have the benefit of a work area designed specifically for the task (raised holding platform, strategically placed processing bins and comfortable seating would have helped greatly} nor was the investigator particularly well qualified to operate the machine in use. Without detailed field data from an established operation, there is no way to evaluate the effects FIGURE 5 41 Measured thickness Micrometer face Non constant thickness measurement error due to surface irregularities 42 f any of these parameters. The figures arrived at, however, can e useful as long as their inherent weaknesses are understood. The 1/4" drill bit performed well in a majority of cases, though t shattered some shells and set up serious stress fracture patterns n many others. These stress lines appeared as thin white lines preading from the point of contact and were visible to the naked eye. An attempt was made to evaluate s>tress fracture patterns using [-rays from a standard hospital X-ray machine. This attempt failed •apparently due to the very small differential thicknesses involved. The process of drilling the hole seemed to involve two steps. In the first stage, the base of the bit actually cut or abraded ±e shell, widening the hole to the outside diameter of the bit shank. Sxcept for very thin shells, there were few failures at this stage. The second stage was entered after the drill shank broke through, :ausing the majority of failures. The unsharpened edges of the shank appeared to catch the interior edges of the hole, tearing away large chunks of material and causing significant stresses. The result was aither large jagged holes or fractured or shattered shells. The possibility of using a larger drill was considered as a means of reducing or eliminating the fractures attributable to shank break-through. A 7/8" drill was run at 476 rpm and depth limited to prevent the entire face from penetrating the shell. This provided a means of using only the cutting face and not the shank. Entry was attempted from both sides of the shell. The preliminary results were very encouraging — losses dropped to approximately 30%. The quality of the hole also dropped, however with a large number being very jagged or seriously stressed. The / 43 :echnique's success was also a function of time spent per shell. As processing rate increased, so did shell fracture. The data present- id in Table (.5 ) was the optimum speed from the results (highest suc- :es per unit time). At this speed, (9.3 holes per minute), the cost .s too high. The final effort was directed at reducing the hole size, thus .imiting the effect of the drill on the shell. Working on the assump- :ion that a vinyl coated wire (O.D. .1") is an economically sound proposition as a string material, a hole size of 1/8" would be suf- ficient to accomodate it. Tests were run first on single she]. Is at .ow speed (476 rpm) . The results were encouraging enough to warrant further tests. Drill speed was raised -to 1338 rpm, improving results slightly, as shown in Table ( 5 ) . The processing time was also re- luced. Group processing was attempted as a possible means of speed- .ng up the process still more. The shells were scattered on the )ench as before, but prior to drilling were stacked in groups of 1-6 and drilled simultaneously. Stacking time was included in the let processing time. Only convex entry was attempted. A maximum speed of 25.7 successful holes per minute was possible ising this technique. Approximately half of the shells rejected were processed successfully except for; an offset of the hole. Many of the land punched shells were simularly mispunched, but were strung anyway without any apparent consequence. Considering those shells with displaced holes as being acceptable, the successful processing rate exceeded 28.7 holes per minute. Since there are no data available on the current practice of land punching the holes, no efficiency or cost comparisons can be 44 lade. The figure of $.02 per ren for punching presented in vppendix II seems to be quite low. This figure represents a success ate of 26.7 holes per minute, based on a $1.60 hourly wage. A more •easonable figure would probably be in the range of 15-20 and may ie as low as 10-15 without presorting. On this basis, it is safe :o conclude that drilling has a definite potential for increasing ihell reliability and decreasing cost.. 45 FIGURE 6 o in 20 H 15 fd -P O -P 10 i 1 / i \ / / I / l n i r • / - ■ v / ■ ' 36 .40 44 48 52 56 60 64 68 .Shell thickness (thousanths of inches) SHELL THICKNESS DISTRIBUTION T3 0) en w Q) u 0 u Q* >i iH rH -p 3 c 4-1 6 * \ ctf tr ro \ ^-^ CO 4J • • (0 ro CO CU r-- CN CO, • CU 0) ro VO cu 00 u o o CN o I ro CO <-\ VO o w j » — W O cA° • ■ • • <: Eh ■J. <*> CN r- Eh to 1 CTi o CO in g 10 • f-i CN • in CN • s CO CO H 2 1 1 1 rH CO u 3 1 z ro in ro rH •*r X U in o "<* VD *r CO <*> <*> u> *x> vo <£> in Oh • • • • • • CCJ <-i in CTl 00 M u * 04 • H CN o rH CTi CN CO O K in in m ■^Jt in J > H r3 < CO w CN H CN CN in as CO in CN m CN H ro Is* m ro CT» CO ro en o u & a, cu N K ro co ro co <^> * • • • o CO to ■H o\o CO u dp •^r T CN r» CO ro VD CO U cn • 00 • * 1 1 1 1 r* CO • ro • r-1 r^ ^r CU rs D in in a * 1 1 1 1 1 1 1 1 • CO a o\o CN ro m CN CN O in m o ro cu d) T3 CU W fi CU ►h CO CU m CO < ^ in CTi V£> CTi CN CTt \D m -C H a • • • • CO • • • • Cn 5 « <* CN rH m o ro CO CN t ro CN CN r* •H 0 iH TT f-i in r-i <*>i- X! rH cs CO II 1 cu a •** T i c9 OP o3 • • • CO CO 0 OP CO a\ cr> rH V£> CTi h> o\c CN CO CN O <-^ r^ • • u 1 ro i u Eu OP • • -•• • X o o CTl ■^r • u ■p vo ro «£> O O Ph C* CN r j r-i r^ OQ rH O CN ■p c cu 0) •H c CU cu •H CO c 0 •H ■p •rl CO 0 H H H X > X > X X > X > X > CO cu s ■ £ z z z B r z z r CU CO a CU 00 CO 00 00 00 *r TT 00 00 T> «# •*r X > •H 3 3 \ \ \ \ \ \ \ \ \ CU \ \ C 0 o O rH rH u CJ CJ3 CO CO CO CO CO CO • CO • CO • CO • CO CS CO • CO o II O 1 CO 47 NOTES OF TABLE C 5 ) F Fractured (See fig. 8-A ) SS Seriously stressed (See f is . 7-D ) » S&MS Stressed or minor stressed (See fig. 7-B ) R&J Rough or jagged (See fig. _ ) MP Misplaced F/S Fractured or shattered (See fig. 8-D ) FP Failed under pressure (post processing testing) C Weak center (See fig. 8-C ) NA Not acceptable * Figures enclosed in parenthesis indicate values if misplaced holes were considered acceptable. 46 FIGURE 7 41 FIGURE 8 So FIGURE 9 a) O Xi V:*g;-f w-.-ti*** nr»-** 51 APPENDIX IV Scallop shells were broken by hand into small (1/4" X 1/4") nieces and rolled in a 1 gallon ball mill for several hours . The 9 residue was sifted through a #20 mesh sieve (.0328" opening) and then :hrough a #8 mesh sieve (.093" opening) to separate powder, acceptable jrit and unground pieces. The powder was used as filler and the grit /as reserved for surface coating. Before constructing test disks, it was first necessary to de- :ermine the best method of manufacturing them. Hot compression nolding using powdered polypropylene was decided upon, principiy because of the availability of apparatus. Molds for the initial test bars were milled from .5" brass stock to provide a 6" X 2" X .125" inset. Later molds were cut on a lathe from .5" aluminum stock in the shape of a 5" diameter disk and later a 3.5" diameter disk. All molding surfaces were coated with Frekote 33 mold release solution to facilitate removal of the molded material. Two types of polypropylene were tried as binders: Hercules rade 6523, a pellet form with a tensile yield strength of 4900 psi and Hercules grade PC 072, a special grade flake with 30% glass fibre and a yield strength of 15,000psi. The powdered grade PC 072 *as found to be superior for the type of molding employed. It formed a more compact, stronger cultch with ^ore shell on the surface than the pellet form. This was probably more a function of the molding technique than the material itself. The flake grade PC 072 was used in all experiments reported here. Polypropylene was dry mixed with powdered scallop shell on a ■ 52 srcentage weight basis and poured into the molds, which were pre- dated to 450°F. Powdered scallop shell was used as a filler pri- irily because of its availability, as a necessary byproduct of the ill mill processing for surface grit. It is anticipated that in jmmercial production more efficient means of grinding the shell )uld be utilized, and that potash, or some similar low cost, low snsity material would be used as a filler. The result would be a Lghter, cheaper cultch. Various filler concentrations were tried, starting with 30% Lller, 70% polypropylene and going as high as 65% filler. While > quantitative measurements were madu, qualitative judgements based l flexibility and fracture strength indicated a 50-50 mix was the Lghest filler concentration consistent with strength requirements. After the mixture was heated in the mold for a minute, to bring t near flow temperature, the press was closed and subjected to 80psi Dr a period of 7-9 minutes. Excess plastic flowed out between the Did and the top plate. After heating, the pressure was released, id the mold placed in a tub of cold water. Removal of the molded Lsk took place as soon as it was cool enough to handle. Releasing le pressure while the plastic was still at flow temperature allowed 2-expansion, resulting in a plate thickness 30-60% larger than de- igned for. This also had the effect of decreasing the density and acreasing the porosity of the disk. Two methods were tried for embedding the surface grit into the ace of the disk, differing only in the point at which the surface cit was attached. In the first, referred to in the future as the -step method, a thin layer of shell was laid down in the mold before 53 Jie plastic mixture was added. The 2-step method required that sep- rate manuf ac cure of a plastic filler disk which was cooled and daced over a, thick bed of heated shell gut. Pressure was applied or 2-3 minutes to imbed the shell in the lower surface. The first prototype molds were 5" diameter disks milled to a 1" depth. Six prototypes were eventually produced from these molds .nd later used in abrasion and selectivity tests. It was felt that t smaller, thinner disk would be better, hence two 3.5" disks were dlled to a depth of .05". The argu3ment for the smaller disk is an obscure one. There is lisagreement as to the characteristic dimension relating cultch size :o oyster production. Results from Marshall would indicate that jetting area characterizes oyster production. Some commercial grow- ;rs, 'however, feel that the setting perimeter, not area, is the limit- .ng dimension. That is, that the oysters grow radially, requiring a ;et amount of arc length for uncrowded growth, regardless of the radial length. While this has not been confirmed by the scientific :ommuntiy, it is worthy of note. A more obvious reason for reducing the cultch size is to prevent ren interaction. The spacing between rens is around 6-8", hence a 5" dish beaves little room for sway. The figure 3.5" was arrived at >>y considering another of Marshall's findings. His results clearly .ndicate that the most desirable spacing between disks is approximately >". Since the ren length is restricted by rack size, increasing the spacing requires decreasing the number of disks. By increasing the spacing size to 4", 15 disks can be strung. This requires a diameter )f 3.49" to conserve setting area, hence the 3.5" diameter chosen. 54 Jsing r as the characteristic dimension, instead of r (perimeter, rather than area), a net loss of 7.5" per string or -12%, will re- sult from this arrangement. r,*> 10 t~ — 'ii 1 4m.1t 1 misMMPi1""' rt ti- ft h ro fD fD 01 *d T3 ^ c Bi (?> ei 01 fD HHHDi 1 | oi w w ft rt- rt rt 0 (D fD o *0 T3 'O TJ fD WO\Oh o en C^P Hi <#> d(> fD Hi O Hi Hi H- ft H- H- I-1 M H H Hi »-■ M fD H- (D fD H H H H M "* ^ fD H vQ*0 H O O h-s: 0 ft a 3 fD o 01 H fD c 3 H 01 ft Hi C h $» h &> O Hi ft (D 0> H- o 0 fD 3 (D 3 & 3 0 M Qi H- 3 vQ ft fD 0 3* 3 H- C fD 01 «)S 57 FIGURE 12 * .v: ^i;,rjiW " ■'■■■■ ■.. ,,-'■.; v,.