�03! 471-1310

PACIFiC SAUNON AQUACULTURE PROGRAM

Incubation and Cultivation Phaaea

Not available for distribution

Sea Grant Final Report

Principal Investigator; John M, Lindbergh

say 1 1 8g

IHIP '»a>i

QzCUQHNQ CÃt

Qa Grant GeNIOSitOq

Ocean Systems, lnc.Domsea Farms, Inc.

11440 Isaac Newton Square, NorthReston, Vtrgrnia 22091

,";aeniggtogL Sea Gx'~tDivision oS Marine Resources37l6 Brooklyn'Avenue S.E. HG-30Univereity oX %aehiagtonSeattle, Iaehiagtan 98195

TAB LE OF CO.'vTE4TSPAGE

ISLMXIAR Y

0 INTRODL.'CTIi".i4

2. 0 GKX'ERA I. DF:5 RII'I'ION VF F'ROJEC I'Z. I I>ackqri> hand

2, 2 Oh~ective

2. 3 General Technical Approach3. 0 DESCR IPTION OF' FROC3RAM

3. I Incubation Phase

3. 2 Freshwater Phase

3. 3 Saltwater Phase

3. 4 Harvest and Marketing Phase

3. 5 Publl c Awareness

3. 6 Environmental Monitor ing

IQ

41

47

~ ~n of gg g 9 ]$73Jgarha Rotourcea

4. 0 CONC LI:SIONS, DISCUSSION, AiVD R ECOMMKNDA TIONS 50

C

fJ

0 Q0

4 000

I'Ii< I, S, Departr»crit of Commcr< e ~v arded Sea Grant bio. P.OH- i <! 4 <-< ii Ii to Oci a» SI,stenos, Inc. for the perineal ApriI I, 1971 to DecemI>erI »7 I to partially s»pport the worI descriI>eri in this report. Subseq»e»tto tne terr»inatio» of the Sea Grant, Dont sea Farms, Inc. was formc d

as a s»I!sidiary of I »ion Carbide to en" ave in salmon aquaculture as abusiness pr oiect. I he total aquaculture activity formerly withQi ean Systen~s, Inc ., v as purchased by Dornsea Farxns and continued.

The period covered by this report through December 31, 1971,describes the program while under the supervision of Ocean Systems, Inc.The period after December 31, 1971 describes the program whileu»der the s upe r vi s ion of Dorns ea Fa rrns, Inc, In order to avoid conf us ionto the reader, the name of Ocean Systems, Inc. is used throughout thereport to represent both the contributions of Ocean Systems as well asDornsea Farms, Inc.

The principal lnvestigat or for the program throughout its entiretywas her. Son M. Lindbergh

C, D. Z.erbyPresident

Dom s ea Far rn s, Inc.

SUMiViAR Y

This is the final report of a Sea Grant project to demonstrate thefeasibility of raI.sing marketable pan-size salmon on a commercialscale in Puget Sound, Washington.

8 total of Zl I, 000 salmon were successfully r aised from eggs to anaverage marketable 1Z-inch size approximately 340 grams! in fourteenmonths using floating net pens. The following phases of the operationare described in this report:

l. Incubation of eggs2. Freshwater c itivation in ponds3. Saltwater cultivation in net pens4. Stimulation of public awarenes s

5. Environmental monitoring

The program was carried out u~der a cooperative agreement betweenOcean Systems, Inc. and the National Marine Fisheries Service NMFS!.

1. 0 INTRODUCTION

In the Pacific Northwest, the salmon is the most highly prized ofall food fish. Pacific salmon resources have been depleted by rnan-made alterations of rivers and estuaries and by the effects of fishing,To offset the depletion of salmon stocks, an elaborate system ofhatcheries has been established. Although these are highly refined systems,they deal only with the early stages of the life cycle of the salmon. Inan intensive system of salmon cultu. e, the fish would be reared throughtheir entire life cycle. It would then be possible to apply a. full range ofmethods of husbandry and selective breeding to the development of stocksbest fitted to the demands of the market.

The availability of fresh salmon is seasonal. A culture system thatwould permit control of the quantity, quality, and time of harvest of stockswould have market advantages. In addition, the early harvest of the salmonwould permit the introduction of a pan-sire salmon of approximatelyIZ-ounce �40 grams! size. A small portion-sized saIrnon is in demandby the ma r ke t.

This report describes such a project to raise salmon on a cornrnercialscale. The program is based on experiments carried out by the NationalMarine Fisheries Service Biological Laboratory, the Washington StateDepartment of Fisheries, Dr. Lauren R. Donaldson of the University ofWa shington, and o the r s.

P.. 0 GENERAL DESCRIPTION OF' THE PROJECT

This project demonstrated the feasibility of a cornrnercial- sizedpilot operation to raise pan-size salmon from egg to market size inenclosures in Puget Sound, >washington.

Ocean Systems, !nc. prov ided rnanagernent for the direction andcompletion of the program. In cooperation with the firm, the NationalMarine Fisheries Service Biological Laboratory, Seattle, providedtechnical advisory services and facilities. The project was conducted incooperation with the National Marine Fisheries Service, the WashingtonState Department of Fisheries, and the University of Washington. Thesite was leased by NMFS from the Washington State Department ofNatural Resources and was made accessible through land owned by theU. S. Vavy and the Environmental Protection Agency. A permit to installthe pens was obtained from the V. S. Army Corps of Engineers. Sixmonths after the project started, matching support for the saltwatercultivation phase was received from the National Sea. Grant Program.

Wa.shington State's first salmon hatchery was built in l895 on theKalama River and was ab!e to release 1, 000, 000 fry each year. Sincethen, the hatchery system has changed into a combined hatching-rearingprogram capable of releasing 100, 000, 000 fry and migratory-sizedsrnolts into the wild environment.

Fifteen hatcheries are located on Puget Sound tributary waters,Willapa and Grays Harbor waters, and eight are loca.ted on the ColumbiaRiver tributaries.

Salmon have several advantages for culture over other fish. Theyare highly prized as food, they at e fast growing, and their hatcherytechnology is the most highly developed of any marine fish. The factthat five species of Pacific salmon and numerous races exist makes thepossibility of genetic selection promising. 1

The Puget Sound area is blessed with an abundance of marine resourceswhich are now used in only a limited way. Intensive fish culture wouldutilize the large supply of clean flowing salt water to produce a supply ofhigh-quality food fish beyond that which is now available throughtraditional fisheries. This would not only provide a non-polluting industryand jobs for the area, but would also compete with imported fish products,and support auxilia,ry businesses such as fish food manufacturing, fishprocessing, and distribution operations.

Lauren R. Donaldson, 'Selective Breeding of Salmonoid Fishes,"Contribution 4135, Conference on Game Agriculture, Oregon StateVniversity, Marine Science Center, Newport, Oregon, May 23-24, 1968.

Z. Z ~01'ective

The overall objective of the program was to stimulate aquaculturedevelopment by demonstrating the feasibility of commercial scaleproduction of pan-size salmon.

?. 3 General Technical A roa.ch

The program described in this report consisted of six phases:

Incubation of salmon eggsCultivation of fingerling salmon in freshwater pondsCultivation of salmon to a. marketable size in floatingpens in salt waterHarvesting and test marketingStimulation of public awareness of the aquaculturepotential in the Puget Sound areaEnv ir onm en tal moni to r in g

4.

5.

6.

The description of the first two phases mntains little new informationabout salmon cultivation. Techniques which were employed are consideredstandard and the information about these phases is included for completeness.

l he S«l>n LLL i-:,it,>ring prf>-'r.>III CIJnS S e<I 'I' f ! ive naaJnr pha;< 9,inst>lvinu in< i>l>atit~>i,>f e > s, jrexhvhater c .lti vatic>n, SaltWater <,Llt>h ~t>tn!arket nq «i>d nti»>i»atittn >f i»tl>li<. awarnneSS. The following sl L t>ttn-W'ill dt C riile th6 t'!> i,.<-t>V< s, i>r<><ediirl s, and reaiiltS Of ea<h pha..e.

I inc lb.l't li > i; f>ha s <.

The art>f C>a'l sl>av,n>n>< >f salmOn egg> s, their subSequent inc> L>a«t>nand hatching, are ac<-omplished roiltinely by many state and federal hat<i> 'r.s-in the Pacific iNo rthv, est. Althu«gh specific techniqiles and facilit>c s va r y,the procedures are well established and constitute no new problems.

The objective of this first phase of the pilot opez'ation was to providesufficient salmon fry for silhseqilent c.iltivation. The hatchery techniqiie swere similar to those described elsewhere, Events included egg and spern>collection, inclibat ion, sorting, and hatching. Of 700, 000 coho egg sp >r chased, 420, 000 l f>0~p! were hatched. Ln addition to the coho, approx-mately 400, 000 chinook fz y were later included in the project. The chinookeggs were incubated and hatched at the Vniversity of Washington.

3. l. l Facilities Descri tion

The coho egg inclibation phase took place at Beaver Creek, a smallfreshwater strearr> flowing into Clam Bay at lvlanchester, Washington. Fig>>re L!- Authorization was given by the V. S. Navy to convert a~n>g>LL surplus b >ilding I Bldg. »7! next to Beaver Creek into a temporaryhatchery LFLgures Z to SL. Nine i6-tray Heath Techna vertical >lowincuhators iviodel t>>691 were installed according to the ptas shown inFigure 5. All plurr>bing was of PVC, exposed wood was coated withpolyester resin, a.nd steel was coated with asphaltum varnish. Four2, 000 kilowatt glass heaters were installed in the downstream side of theheader box. With straight-through flow, temperature in the incubatorsheld at about l. 5oC above cz'eek temperature. A recycling system and filterv,ez'e installed and placed in operation on Z5 iVoveznber. Filter and bypassvalves were adjusted to provide for approxirrtately 80<0 recycled water, whichallowed temperature to be maintained at 5 C above creek temperat >re.Two ther mo stats, each controlling two heater s ~ wer e set to allow a maximumtempez atuz'e of l l o C Due to construction delays, water flowed throughthe system for only two days prior to introduction of the eggs. Eachincubator received approximately 3 to 5 gallons per minute flow. Theresults of water analyses at Beaver Creek during this period are shownin Table l.

3. 3.2

A permit was received from the Washington State Department ofFisher ie s to purcha se oo, 000 coho salmon eggs from surplus SkykomishQat che z y stock Z hi s hate hery is located on May Creek, about four mile s

-5-

LFigure I

6

The incubator facility atHeave. r Creek. The sri.all

building to the left is theput!ip hoiise. The headerbox is niuunted on the roof.Inciibator stacks are moiinted

inside the building.

FIGUR K 3

Inside the incubator facility.Incubator stacks are at the

right standing on the drainagetables. Laboratory equipment isassembled on the left. Windows

were covered with opaque plasticto prevent light shocking of eggs.

FIGURE 4

A typical try of eggs inHeath incubator. Fiber-

glas s trays ar e s uppo r tedin aluxninum stand and

water cascades throughthe trays.

V

0 20

fil F

g D0Z

0 Q.

X

0 4

Cl a.

O

gp aZ

g 0

4 0

f- 0OU

0

>epU0

A 0mC

a

gEp 0

9

gS

g

Z <

0 0a.H &

Q Z

4 X4 04 0< Z x

U

A D

C

X 0'U

A

gEp, C4p, C4gg

0 Cg

nl0 0

a.F 8

east of Sultan and -10 n.iles not tbeast of Seattle. Qn lv'overnber 6 1970the hatchery, gav e notification that t'ipe coho:almon had arrived at the ueir.Project personnel, supervised by Anthony Aovotny of .'v'MFS subsec,uentl 'uentook eggs on ~ovens!ier 9 and 19 1 Figures o to 14.

Skykoniish fi~tchery eniployees ran a seine net around several hundredfish anti r t uu deA the in against the bank. Small nurnbe rs of fish w ere die ipnetted, naales seg reenter from fertiales, and the la i gest and fittest of eachse le ct e d.

Eggs and sper rri u ere packed in ice and transported to a hatchery atBeaver Cr eek near hlanchester. Each gallon of eggs vvas poured into a.clean five- gallon pla stic bucket and one cup of sperni added. The eggsand sperm «ere allowed to sit for one minute, then they were ~ashed,drained, and split into two incubator trays,

Dead eggs were picked from trays during the period from 18 Decemberthrough Zl December. Ivlortality was considerably higher than anticipated.Eight trays were particularly bad, having mortality rates of 90~a or more.In all cases these trays were paired, indicating a problem with sperxn orfertilization.

On ZZ December the eggs were ueighed and th» trays balanced so eachtray would contain a.pproximately the same number of eggs. Residual eggsfrom trays with severe mortality were kept together. Four one-hundredml samples of eggs were counted from three separate trays and weighedto provide a weight count. Extrapolating this count to the total weightot eggs gave an estimate of 469, 000 eggs surviving from the original700, 000 for a mortality of 33%o.

Extensive hatching took place on January 12-14, and by January 15over 95'%%uo of the eggs were hatched. On January 31, the estimated numberof live coho salmon was 445, 000 or 63. 5 ' of the origina.l number of eggs.

The major problems found in the incubation phase were low watertemperatures. silt deposition and saprolegnis.. The first two of thesecan be corrected by proper engineering design and choice of water supply.Supplemental heating adds significant costs to the operation, while adequatefiltration is relatively inexpensive. The saprolegnia problem can be minimizedby treatment with a fungicide before extensive mortalities occur.

Chinook incubation and hatching was conducted at the University ofWa s hington.

3. Z Freshwater Cultivation

The freshwater phase of the salmon culture pragr am spanned aboutsix months, from the time of hatching until the fingerllngs were largeenough to adapt to salt water. The goal for this period, production af

-IO-

r= IG'R E ii

Seinin ~ of ~ old spav nin~ stocka.t the Skykcmish hatchery i.nV'ashington State. These fishwere returning to the hatcheryv here they were spawned ar.dreared 3 or 4 years ago.

FIGURE 7

Selection of the prime cohospecimens. The fish wereselected for size, shape andfree from disease or defects,

FIGURE 8

After selection, the fishwere killed and bled to

facilitate contaxnination of the

sperm by blood. The fishwould normally die soon afternatural spawning.

Collection of eggs from fen-.-: lec oho salmon. The a bdom. na I

wall was slit and the ripeeggs spill into the collectionpail. A t thi s point it i simportant that slime or bloodi kept away from the eggsas contarr ination will reduce

f e r tili z ation.

FIGURE l0

After the eggs of several fish arecollected, they ~ere decanted into1-gaUon jugs and iced for thejo urney to the incubator.

Fertilization took place at theincubator facility.

12-

>tr sooln+ of sperr,l fromn>ale coho salmon. Spermwas ''rr'ilked'' from the salmon

by ma s s a g i ng the a bdorninalwall. The sperm was collected

in snlall c ups for tran s po rtto the incubator at Beaver

Creek.

FIGURE 12

After delivery to the incubator,the sperm and eggs were xnixed andplaced into incubator trays. Theeggs were left undisturbed ~r -ilafter they had eyed out. At thatpoint the dead and infer tile eg g swere removed.

-13-

FIGURE 13

A typical tray after the eyed out stage. The white eggs are dead o' runfertilized. Black eggs are also dead and have become covered withsaprolegnia a fungus!.

FIGURE 14

Picking of the egg trays. Each defective egg was removed with tweezersand transferred to a graduated measure. The healthy eggs were washedand returned to the incubator until they hatched.

-14-

700, 000 chinook and coho fingerlings, was met during the surnrner months.Major tasks for the period included feeding, cleaning, predator control,temperature and oxygen adjustments and inventory estimations.

3. 2. l Facilities

The freshwater cultivation took place at two freshwater ponds Glud's!near Brownsville, Washington. These ponds total about one acre of area,with an average depth of lO feet, and are fed consecutively from a singlecreek with an estima.ted minimum flow of 500 gprn Figure» !. The inletsand outlets can be controlled to fill or drain the ponds. Each has a mudbottom and dirt banks covered with a veneer of sand. Temperatures andwater analyses taken during this period are shown in Tables II and IIL Toincrease the low water temperatures, the oil-fired heater from the hatcherywas installed at the upper pond and delivered ZO gpm of heated water to thepond with a 15oC. '=fferential. Pond temperature was raised approximatelyloC. The fish, . wever, tended to stay in the vicinity of the discharge,which was significantly higher. Heating continued until mid April whentemperatures rose to over 10oC.

3. 2. 2

Approximately 420, 000 coho fry were transferred froxn the incubatorsat Manchester to Glud's upper pond on February 6 and 7. This wasaccotnplished by truck, with a.n insulated fish transfer box and oxygen,so that very few fish were lost in the process. The fry were placed in4' x 8' x 6' net enclosures, designed to crowd the small fish to help themstart feeding. However, during the first night the small fish moved to thebottom of the net which bagged and caused them to pile up. The fry on theinside of the pile began to suffocate. As soon as this was noticed, the fishwere released to the pond. The survivors immediately schooled and appearedto be feeding adequately. An estimated 300,000 coho survived the transferand abortive crowding attempt.

Subsequent to this incident discussions were held with Dr. L. R. Donaldson,of the University of Washington. Dr. Donaldson agreed to donate 463, 000surplus chinook fry for inclusion in the experiment, in order to providesome oemparative data about the two species. These fish were produced atthe University by genetically selected spawner s froxn a strain that has beendeveloped by Dr. Donaldson for nearly 20 yeaxs. They were selectedfor maximum growth and disease resistance as well as other factors. Apermit was received irom the Washington State Departxnent of Fisheries toauthorize the transfer and the inclusion of chinook salxnon in the experiment.

These chinook fry were transferred to the lower pond in seven loadson February 9-I2. iVet crowding enclosures were not used, and the chinookstarted feeding ixnmediately. A total of 463, 000 fry with a tots.l weight of668 pounds were thus stocked with. very few mortalities.

ym .I e~-3 ~ '

FIGURE 16

Preparation for transfer to fresh-water cultivation. The transfe rbox is mounted on the truck parkedoutside the incubator facility.A standard insulated box witho wyg e n b ubble r s ys tern wasutiliz ed.

Giud' s Pond near Browns~ ilie,

Abashin"ton. The two pondshad been developed for fr eshiiatercultivati.on of trout and had been

u s ed by KV a s hi nato n StateDeI <rtment of Fisheries forsalmon c«ltivation..

FIGURE 17

Transfer of the fry from thetrays to the transfer box. Oncethe fry had. absorbed theiryold sacs, it was necessaryto transfer them frozn the tray

to the pond.

FIG' RE 16

Each tray was carr ied to thetruck and the inner container

was lowered into the transfer

bo x. The s c r een wa s r ernov ed

and the fry poured into the box

FIGURE 19

Temporary crowding facilitiesat Glud's Ponds. The 4' x 8'

nets were intended to serve a.s

low-cost hatchery trashs in whichthe swim-up fry would startfeeding. The crowding facilitywas not successful.

-17-

Clark's ''.4ev. Ag~e ' salmon feed, a dry formulation, was chosen to beginfeedir . both the chinook and coho fry. Economic and handling considerationslead ' . a dry rather than a nioist pellet. Feeding was begun at 6~0 of totalfish w.ei ~ht per day. Qf this a passible 10' o was lost to the pond bottom andsides w bich requires hand raking. Four automatic feeders were installedduring %larch, feeding at l5-minute intervals from dawn to dusk. Supple-mental band fee~iing was reqiiired once each hour to better distributefood to the larger chinook pond. As the fish gr ew they were graduallyshifted to coarser grade pellets.

In mid-%larch it became apparent that the chinook salmon were notgrowing well. Their diet was changed to Clark's Oregon Moist Pellet II,and they resumed normal feeding. The effects of the chinook diet changecan be seen clearly on the growth chart, F'igure 21 and Table V. Thecoho continued to feed normally on the dry diet.

By mid-April the density of fish had inc.reased to the point thatsupplemental aeration was required. A small air compressor was installed,an additional 12-I nch pipe was installed between the ponds and the entirecreek flow directed through the ponds.

In May, rising temperatures and dry weather required installationof additional aerator s and several recirculation pumps. Water flow in thecreek dropped to 550 gpm and dissolved oxygen in the lower pond fellbelow 6. 0 ppm at one point. On May 19, the transfer of chinook to saitwater started, so the oxygen demand in the lower pond was reducedimmediately. In June and July, low dissolved oxygen in the ponds requireda red~ction in the coho feeding rate. Their growth during the projectis shown in Table IV and Figure ZO .

Predator control was attempted throughout the freshwater cultivationphase. Since the ponds had been drained and cleaned prior to stocking thesalmon fry, no native fish were suspected. However, at the conclusion ofthe freshwater phase a total of 66 very fat healthy cutthroat trout wereremoved during the coho inventory. These accounted for a loss of anestimated 33, 000 smaQ coho.

Discs,se was never a problem during the freshwater phase. James W.Wood of the Washington Department of Fisheries examined the fish onApril IZ and found their general health excellent. No diseases or parasiteswere found in many subsequent examinations.

I

3. 3 Saltwater Cultivation

The unique aspects of this aquaculture pro ject concern the saltwatercultivation phase. Introduction to seawater occurred at a sise ofapproximately 5 grams for chinook and 15 grams for coho.

~ 1 ,' ~0 ~

05

agC

gJ,/I ~

fL.S i.g

IO 0

Jt JJ+

a,%f c

.o.e

.ef

OCCaN IVSVee iNC.

IN~NON 14&COXVhf MNCC T

l 970< 872

COHO

444 fCO 1%0

ataunt 20I9

O

5IL

o>

Xgy IIek

P'A 51I AP1 SAY W%C 4Vlf 4UO OM OCT 0% RC

0, I

1r

TABLE IV

COHO FEEDING

Total Keight1st of mo. lbs.

Weight

Increase lbs.

Food

Fed lbs.

Food

Conver sion MONTH

F e br~ia ry

lviarch

April

420 130 300 2. 30

460 1260S50

1250 22401020 l. 79

Z7902270 2800May 1. QO

5060 5090June 4350 0. 90

3750 l. 36July 5090

2300 6200 2. 70

14, 000

Z6, 490

44, 200

14, 050

26, Z90

33, 075

21, 125

I . 00

O. 99

0. 75

8535 2. 30

75, 797 loss! I Z, 600

20, 639 loss! 5196

IZZ4 625S 5. 11

76213072

5685 12 ~ 630 1,98Mav

13, 617 loss! 10, 700Jun~:

July

I Total Weight, weight increase, and food conversions are estimated.

2 Food conversion = weight feed fed/net weight increase of fish

-20-

August

September

October

November

December

January

February

March

Aoril

10, 150

13, 900

16.200

30, 200

57, 390

101, 590

110, 125

34, 328

13, 689

14, 913

17, 985

23, 670

10, 053

c5 oO < o

~0o 0

I ~

4 aC

7. o

C.&

X. d ~. git 4

8. 7

gory 93,c+ta

Cg, les

d,0$

.OI

qO

CMINOOK

fleURK212'!'

I

a

5OA

tl1

X~4

30

OCEAN SV%7%NS INC,

SALi&1 10VACLLTLNE PIIOJECT

I 970-I 972

S 5A1 AM SAY WOC JUlf All ~ t OCT OT WC IAll fthm 8%0

TABLE V

CHINOOK FEEDING

Total W eight1st month lbs.

WeightIncr ea s e lb s.

Food

Fed - lbs.

Food

Conver s ionMOiXTH

668February

Mar ch

April

87 500 5. 75

2. 98

1. 74

755 570 1700

1325 1955 3400

May 3Z80 3ZOO 3950 l. 23

6OOGJune 3160 5940 1. 88

July 9160 3400 7150 2. 10

5340 Z. 11

17, 550

11, 850

1. 10

2. 49

6970 4. 4Z

19, Z41

1G, 971 Z. 15

6. 384848

3920

?. 468520

78, 131 lose!

15, 946 lose!

May

June

7693July

Total Weight, weight increase, and food convereiona are estimated.

2 Food conversion = weight feed fed/net weight increase of fish

-22-

August

Septeznber

October

November

December

January

February

Mar ch

April

12, 560

17, 900

35, 450

47, 300

54, Z70

73, 511

84, 485

89, 330

93,250

101, 770

23, 639

11, Z65

19, 155

29, 5ZO

30, 820

ZS, 5GO

Z3, 550

30, 950

21, 300

20, 956

13, 150

6500

3. 3. I Site Descrr tion

Puget Sound is a narrow estuary lying entirely within the contiguouswaters of the State of Washington ancl is connected to the North PacificOcean via the Straits of Juan de Fuca. The exchange of water in narrowconstricted inland seas such as Puget Sound is generally typified by twocirculation patterns. The first, characteristic of areas where evaporationexceeds precipitation, exhibits seaward flow of dense, highly saline waterat depth. The Mediterranean Sea i.s one such example. The second,typicaL of areas where precipitation exceeds evaporation, exhibits seawardflow of less dense low salinity water at the surface. The latter casetypifies Puget Sound where dense nutrient rich waters enter the basin via.Admiralty Inlet in the north, creating eutrophic conditions siznilar to theBaltic Sea. Primary productivity measurements in Puget Sound betweenAdmiralty Inlet and the iVar rows range from 0, S-S. 0 gzrr carbon fixed/m /day2

with an average annual production of 450 gxn/m2. Certain embayzzzents,due to their shallow nature, heavy stratification and turbidity, probablydo not exceed production for the main body of the Sound on an annual basisor in carbon fixed/rn2/day but znay exceed production in carbon fixed/m3/day.

Clazn Bay, a portion of Puget Sound near Manchester, Washington,possesses an ideal temperature range, good tidal flow, protection froznmost storms, a sandy silt bottom, and very little pollution. The floatingpens for this project were moored on the south side of the Bay, in 30-40feet of water. The average tidal range is about 10 feet with currents ofup to 2 knots. Water analyses taken at this site throughout the period aresummarized in Table VI.

Support facilities used during the saltwater phase, such as laboratoryand personnel space, garages, storage space, and loading platforms, wereprovided by the National Marine Fisheries Services including the RVBROWN BEAR!, the Environmental Protection Agency, and the U. S. Navy.Other facilities used. was a workshop, large capacity freezer, fork-lifttruck, air compressor, and various boats.

3. 3. 2 Facilities De seri tion

The concept of the floating net pen is the key to the viability of thispilot project for salmon cultivation. Floating net pens were developed toutilize the following advantages:

a! Free-flowing tidal water currents are used to bringdissolved oxygen i.nto the pens and to flush outxnetabolic wa stes. Unlike pond cultivation, wastesare thoroughly removed and no pumping power isr equired.

-23-

J

Ipal

LCl

DO

Z

D

p 0

D

D

ru ~ D

0

QQI

O. hJ

CCt«DCX!

CO

e

D D'«

rt1

Ig!

ltd CO~ I

O0

0 0 00

Q C4

I

«««

ce

0

I

00

D

~o

V

«0

~ «IX! P4

M

O

I

h3

F ICiURE 3Z

The R I V B R 0 3 .'i HEAR

tied up at vIanchester. Thevessel serves as the lahorato. yand operating base tor theAM FS Biological Laboratory.Pens are tied up next to theBRO'A' 4 HEAR .

FIGURE Z3

Exp e r iree ntal hexa g o na 1 p e ndeveloped by AMFS for ea,rly releaseexperiments for the sports fishery. Thispen has the capability of being submergedfor bottom cultiva,tion.

Experimental net pens are shown in theforeground. These are typically

used for labora.tory scale experiments.

FIGURE 24

The fir s t p r o to typ e floatingnet pen. The unit was 30' x 30' x15' deep. When this picture wastaken it contained almost

350, 000 chinook smolts.

4! The net pen is resilient and yields v,ithhi'h currentforces. At the time of rnaxirnum current the penis partially collapsed while absorbing the pressure.Although .olume at maximum c rrent may heconsiderably reduced, water flov; and hence dissolvedoxygen in the pen is high so the fish do well despitecrowded conditions.

c! The net pen is a relatively low capital investmentfacility v hen cornpa.red to alternatives such as fixedstructures and diked ponds.

d! The large thermal mass of Puget Sound is utilizedto buffer the temperatures tn the pen. Unlike pondcultivation, temperatures throughout the year wererelatively stable and constant in the net pen enclosures.

e! The net pens have a. low environmental profile becausethey do not require permanent changes in the ecology andcan be moved if necessary.

Sin.ce chinook salmon are capable of adapting to salt water at a muchsmaller size than coho, their transfer began earlier. Preparations for thetransfer were started in April with the construction of an intermediate netpen enclosure. This pen measures 30' x 30' x 15' deep and is fabricatedof 3/8-inch stretch mesh knotless nylon of 126 denier strength. It issupported by floats along the top edge and weighted along the bottomedge.

The growing pens were secured to a moored raft as shown inFigure 25. The 100' x 24' platform was constructed of welded steelcylinders and wood decking. It was extremely sturdy and was builtof available materials rather than to meet an ideal configuration. Launchingand anchoring of the float took pl.ace on July 15. After installation theanchoring system was found to be greatly underdesigned. Muchtemporary and emergency work, lasting se~ eral months, was requiredbefore a stable mooring system was devised.

The four main net pens, each 50' x 50' square and 25' deep weremoored to the two long sides of the main float. A system of counterweights and anchors held them taut at the four corners regardless oftidal fluctuations. The netting, knotless nylon of 7/8-inch mesh �26denier strength!, could be repaired by serving and proved extremelyrugged. Fouling with algae did not constitute a problem below 2 or 3 feet.Heavy mussel sets to approximately 10' became a serious problem atthe onset of the spring of 1972.

-26-

27

Fj gure 27

!i oint, pe .'luati.".g iiei !en.

A predator:iet cui,ers the top orthe per..o thwa.t fish-eatingb rds.

F'iGl RF; 2i

The i aft assembly moored in ClamPay. The sn-.a.ll buildings providestora Je arid shelter, The large roiindbi;oys held taut by counter weightsmaintain a constant tension on the pens. FIGURE 30

Details of the pen installationon the main raft. The small

floats support the aerialpredator net that was installedto keep dogfish from chewinginto the main cultivation pens.The inner pen is similar to theprototype but measures 50' x 50' x25' deep, The square floatin the background supports acounter weight that tensionsthe pen.

30�

During the project the maximum current experienced was approximately3 knots. !viaxirnum wave heights were 2 to 2 I/2 feet. Maximum v inds,corning out of the SW, were 40 to 50 knots. Debris was a problem and logs,branches, and macro algae were frequently fouled in the nets and rigging,

The system of anchors and counterweights was inadequate throughoutthe project and continuous maintenance was required. The system did notprovide a ta.«t net system during extreme tidal fluctuation.

In August, dogfish Squalus acanthias! caused serious damage to thepens shortly after salmon were introduced, Dogfish attacked mortalitieslaying on the bottoms of nets and three or four feet up the sides. Withtheir extremely sharp and abrasive teeth, the dogfish tore dozens of holesup to two feet in diameter while trying to eat mortalities, Several dogfi.shwere found inside the pens and an unknown number of salmon escaped.

Dogfish attacks were prevented by installing an outside net completelyenclosing the 50' x 50' x 25' nets at a distance of 5 feet. This predatornet was 60' x I l0' x 30' deep in dimensions and fabricated of 5-inch stretchmesh gill netting. The concept was highly successful. Dogfish onlyattempt to bite directly at food and do not ha.ve sufficient intelligence totry to open up an entry. The net is damaged incidentally to biting and evena small gap appears sufficient. No attacks were experienced even thoughthe two nets were touching during strong currents.

GiU nettin.g does not appear to be the best material for predator nets.It is not sufficiently strong, and it may entangle desirable wild fish aswell as dogfish.

The aerial predaCor net was designed to keep fish-ea.ting birds fromattacking the salmon, but caused chronic problems. The geometry of themain nets became distorted by strong currents and allowed the aerial netsto dip close or into the ~ater. Salmon became entangled and killed, sorne-ti~es as many as l0 or ZO per day. When aerial nets were allowed in thewater for only a short time, algae became attached and the nets were heavyand difficult to relift. Ripe in the net were frequent and extensive.

Seven foot high masts were installed on the main float and the aerialnets were attached to the tops with counterweights to keep the netting ta.ut.This method, while an improvement, was not adequate. The birds werelimited in their access as long as the nets were reasonably intact.

The net pens were supported by canister floats which worked well forthe early months but deteriorated before the end of the project. The floats with plastic pipe risers! abraded and ripped holes in the aerial nets and had atendency to pull free from the headlines. In several cases they rippednetting as they broke free, Under the weight of bay mussels in the springof I 972, the floats began to sink. Eventually the canister floats were

.ntersar cetails of the

cultivating pen, The extendedplanks facilitate access by dxver s.

FIGURE 32

Net pen in strong windand current.

FIGURE 33

Pneumatic feed blower house on

the znain raft assembly. Agasoline powered blower wasused to disperse the feedinto each pen.

32-

replaced by eight-foot long styrofoam biilets that v ere sewn cortinuo..slyaround the skirt pertmeter.

Some difficulties were experienced when the chain weights beneaththe ~r owing pen became tangled in the under water predator net. Theproblem v as solved by retensioning the predator net, but future installationsshould have all weights attached outside the outer n~ost net.

In the saltv'ate r phase, the rapid growth of the fish Figures 20 andrequired ever increasing quantities of food, and this in itself created

new logistic problems. An amphibious truck, DL'KW, was acquired forhauling feed to the main float. It has a ca.pacity of 5, 000 pounds of feed,and permits shore loading from a forklift truck F'igure 32 !. A freezercapable of handling 20,000 pounds of moist feed was installed. Feed v asat first dispersed by hand, but in October an air blower was installed onthe main float in an attempt to increase efficiency.

3. 3. 3

The first chinook smoits were transferred to salt water on May 19,1971. No prophylaxis treatment against disease was attempted. Smoltswere seined and hauled by truck to Manchester, where they were briefiyadapted to the salinity and temperature difference. Salt water wasintroduced into the tank over a 30-minute period before the fish were drainedthrough a 2" hose 100' long into the net enclosure.

A total of 394, OGG �,000 lbs. ! of chinook were moved to the 30' x 30'pen in 35 trips. About 7, 000 fish were released in compliance with theState permit that aUowed Ocean Systems to receive these fish fromDr. Lauren R. Donaldson. Another l, OGO fish were flushed into thecreek when the pond was drained. Only about 900 chinook �,25fo werelost in tra.nsportation to Manchester.

In early June the first evidence of an epidemic of Vibrio angularans2occurred. Mortalities increased dramatically on each succeeding day.The presence of Vibrio was confirmed by culture analysis. Vibrio is abacterial infection endemic in salt water. It usually affects fish moreseverely in high temperatures or when they are in a weakened state,such as after handling.

On June 9, the chinook were treated with food fortified withTerramycin TM-50!, which was continued for five days. On June 12,the mortality was reversed and by June 19 the attack appeared controlled.Total mortality resulting from the epidemic was about 19, 500,

Subsequent references to Vibrio describe Vibrio *ngularans, notVibrio parahemoliticus.

-33-

F' IG',.'R F 34

2'ning t'rory the feed blov'er

ho»se to each pen. Also shov~are evperimental pens suspendedbelow the main ratt,

FIGURE 35

An amphibious DUKW that wasused to transfer feed from the

storage site to the main raftassembly. The vehicle iscapable of 5, 000 pounds pa;-load pertrip

FIGURE 36

Seining of the fish at Glud si'onds. After the fish v er e

crowded they were bra<ledinto hiickets and rien throi h

a .iielson grader. Variousholding pers and pools areshown wl.ich were used to

prepare 4atches for the truck,

FIGURE 37

The iVielsen grader in operation. Eachfish slides down between slightly divergentbars and falls through into one of threechute s. The smallest wer e r eturned to

the pond; the other two groups weretransferred to the floating pens atManchester.

� 35�

FlGLRE 3S

Transfer to the floating net pen. A truck with smolts is located onthe pier. A pump on the small raft pumps salt water to the box for acontrolled temperature and salinity transition. Thereafter, the fishare drained down the long black pipe into the pen. Shown below is thedischarge with a small chinook srrlolt about to enter Puget Sound.

On June Z9, however, a second outbreak occurred. It wastreated the same *s the first and was controlled by July 8, Themortality from the second outbreak was 25, 800 giving a total of 44, 300from Vibrio in a month. Experience in this project and elsewhereindicates that prophylaxis treatxnent for Vibrio should be undertaken a.ttransfer to salt water and other situations of unusual stress to the fish.

Between July 23 and August 12, the chinook were weighed and thensorted with a Vieilson automatic fish grader. They were transferred tothe two 50' pens on the west side of the main float after Terramycintreatznent. Very little disease was encountered, but scale loss due tohandling, aggravated by high water temperatures, above 14oC! contributedto 33, 63Z chinook mortalities.

Between July 7 and August 31, the coho salmon were graded, trans-ported, introduced to salt water, and stocked in the two 50' pens on theeast side of the float. Only coho larger than 15 grams each were chosen.A total of Z61, 000 coho were successfully transferred with a mortalityof l. 1'%%uo during the process. These coho were treated with Terramycinprior to the xnove and no Vibrio outbreak occurred.

The coho salmon reznained on the Nelson "Silver Cup" dry feed.For test purposes, half the chinook sa.lznon were continued on Oregon MoistPellett II while the other half were changed to Silver Cup dry feed inAugust. The salmon diet figures are summarized in Tables IV and V.The conversion rates shown, while important to the economics of theproject, can be misleading since they include mortalities, losses and attiznes questionable inventory statistics. Their optimization depends onan accurate estimate of the total nuxnber and weight of fish fed. Sinceinventories during the project varied in their accuracy, feeding rates werenot always optimal.

Natural food brought to the nets by currents may have contributedsome small portion to the conversion rates, Preliminary small-scaleexperiments conducted with underwater lights by NMFS indicate that thiscontribution could be no znore than 10'%. Because of the large numberof salmon held in the floating pens, the actual contribution must have beenconsiderably le ss.

Disease losses during the saltwater phase constituted about 33%of the chinook and lifo of the coho. Disease monitoring was perforznedregularly by Anthony Novotny of NMFS, particularly for Vibrio,fux'unculosis and kidney disease. No external parasites caused. epidemics,although an occasional Argulus was found. In order to assess diseaseeffects as well as to adjust the inventory, xnortalities were removed twiceweekly by divers. A pro-rated mortality count greater than 1'% per monthwa s c ons ider ed s us picio us.

-37-

FIGURE 39

Feeding time on the main raft.These photos show the problemsassociated with the aerial pre-dator nets. Various devices

from poles to beachballs wereused to keep the nets out of thewater.

-38-

Growth was ~ ery good durin. Septensber as the 12, oC temperaturewas close to optimum for both species, Both groups doubled in sizewith excellent feed conversions. Disea.se losses were normal in

September, w ith only Vibrio being of any consequence. The chinookindicated a Vibrio mortality rise in mid-September, but were successfullyt r ea ted with Te r r am yc in.

During September, the second group of coho w'ere graded irto twogroups and l2, 250 fish were transferred to VMFS for a feeding experiment.Very little scale loss was encountered when handling the coho in saltwater and mortality was minimal as opposed to chinook which exhibitedvery loose scales in salt water and suffered high losses!,

During October temperatures dropped and began to affect all fish.The coho continued their growth, although rations were cut slightly tocompensate for the lower rnetabolisxn at the colder temperatures. Feedconversions remained constant for the coho and there was no disease.

Both pens of chinook once again contracted Vibrio and were shiftedto medicated feed. The fish were, however, reluctant to eat it. OnOctober 25, at the termination of treatment, the second pen of chinookwas shifted to Nelson dry food. Both groups of chinook had been growingand converting about equally, and dry food is cheaper a.nd easier to handle.At the end of October, however, sub samples of chinook in both pensshowed sharp declines in growth and conversion. This was attributedto disease and unpalatable medication.

Difficulty became apparent in obtaining reliable sample weights.Dip nets and towed nets up to four feet in diameter allowed the largerfish to avoid capture more easily than smaller fish. Obviously, suchsamples are biased and all resultant plans and statistics are unreliable.Et was found necessary to use a large seine which would sweep a bigarea of the pen.

During November saltwater temperatures decreased to a low of 9. 4 C,Chinook growth rates and feed conversions continued poor. The cohocontinued feeding well and both growth and conversions were excellent.Disease presented no major problems for either species this month.

During December the water temperatures dropped below SoCconsistently and both coho and chinook began feeding poorly. To attemptto stimulate feeding, the chinook in the southwest pen were switchedback to Oregon Moist Pellets but little difference was noted betweenthe moist and dry feed. A surprising,but not serious, mortality wassustained in the first pen of coho approaching market size. It is suspectedthat the increased density in this pen, approximately l. 6 lbs/ft3, andresultant low dissolved oxygen were major factors.

- 39-

Harvest operations began during mid-December and are described ina later sectio". The largest fish were retained for brood stock purposes.During the last week of December, a rip occurred in the corner of thesoutheast coho pen and an estimated 70, 000 market-size coho escaped.Many of the fish rerriained in the vicinity of the pilot project raft and some2, 000 v ere r ecovered. The bulk of the escapees were classified as a.ninvoluntary release for the Washington State sportsmen.

During January, water temperatures dipped to 5. 5 C and rationsfor both species were cut to 1~a of body weight per day. No significantgrowth occurred. Growth and conversion data for the coho during januarybecame confused as the population was reduced and fish were stressed.Disease was not a problem in January.

Temperatures began to increase during February and consistentlyhovered between 6. 5oC and 7. 0 C. Rainfall and turbidity were unseasonablyhigh and feeding behavior was correspondingly poor on severa.l days,Feed rations were increased in an attempt to stimulate growth on therising temperatures, but no increase was noticeable. Once agai~ diseasewas not a problem and no medicated feeds were required. Harvest of themarket -size coho was completed and a sub market-size group was leftfor further growth. Feed conversion for the coho during Februa.ry isnot meaningful due to the stress of harvesting and transfers.

Warmer air temperatures and gusty southwest winds predominatedin March and water temperatures climbed to 7. 9 C. The chinookcontinued to feed and convert poorly and both species were treatedwith TM-50 due to the rising incidence of Vibrio. No meaningful samplesof the largest coho kept for brood stock were taken. but there wasevidence that they grew significantly all winter.

Water temperatures responded to generally fair and warm weatherin April. Water temperatures increased to 8. 6 C and the increased photoperiod stimulated all fish to feed vigorously. Growth rates and feedconversions increased. The brood stock coho reached an average weightof 2. 5 pounds by the end of April. The chinook finally reached marketsize at approximately 3/4 pounds a,nd harvesting began on April 25.

Both dry and moist diets evidenced marked improvement in feedconversions and acceptability this month with increased temperature.Neither feed appears to be of appreciable growth value belo~ 7oC anddry feed was poor below 7. 5 C

Approximately 3,000 small fish were turned over to the NMFSresearchers for feed and ration studies. The intent of this work is toinvestigate feed conversions and growth ra.tes resulting from variousrations at increasing temperatures and photo period,s.

-40

3. 4 Harvest and hiarketin

The purpose of the project was to demonstrate the feasibility ofcommercial scale production of pan-size salmon. In addition tocultivating salmon, it was necessary to harvest, process, and sell theproduct. Several approaches were investigated and test marketing wasconducted prior to actual harvesting,

3. 4. 1 Descri tion of E ui ment s.nd Process

In October 1971, a trial batch of pan- size salmon was processed byan established salznon packer in the Seattle area. The results and timestudies indicated that the facilities and personnel used to dress largesalznon could not process the pan-size fish economically. A processingsystezn was designed specifics.lly for the pan- size salznon. Availablecornznercial facilities and equipznent were reviewed and a. material balancewas conducted on the system in order to miniznize handling. Materialsand equtpment were ordered for mid-December delivery.

3. 4.2

Fish were crowded within the pens with a 30' x 60' seine. Oncecrowded they were pumped with a six-inch fish puznp frozn the seizie acrossa dewatering device and spiUed into an adjustable bar grader where t'heywere separated into three sizes submarket, znarket, and brood stock!.The znsrket size fish were discharged into holding nets where they wereheId until a full-day's processing run had been accumulated.

Slaughtering by chill killing was selected because it zninimizeshandling, provides the fx eshest product, and elixninates struggling priorto death. Electric shock and suffocation were also investigated but wererejected as inferior znethods. The fish were pumped from the holdingnet across a dewatering device, and discharged directly into a converteddairy cream storage tank containing -I. 5 C concentrated brine. Theywere instantly killed and within minutes their body temperature was loweredto O. OoC. The chilled fish were loaded into 100-pound tote boxes and truckedto the processing plant.

3.4. 3

Butchering and packaging operations were standard and manual,although special care was exercised to znaintain the high quality of theproduct. Butchering operations were physically separated fxom theinspection, weighing and packaging stations. Most of the fish wex e xnarketed"dressed heads on", with only the entrails, gills, and kidney tissue removed..Weight loss in processing was a.bout 20%%uo. A vacuum eviscerator was usedto rexnove the kidney tissues without scraping or damaging the bones andznembranes of the belly cavity.

After inspection, the fish were held in chilled brine uzxtil they werepackaged. All fish were gx'aded into 6-8, 8-10, 10-12 and IZ-I4 ounce

-41-

categories. Each fishsea e . e poucheci fish wl d. Th

as placed into a polyetheiene pouch and heath w«e placed into five-pound waxed cardboardca.rtons which were overwroverwrapped with polypropylene film.

The fish wer e frozer o-40 G in a conventional blast freezer,then packaged into tlirt ty Pound corrugated cardboard niaster cartonsprior to being transferreer red to the wareiiouse, A coding system wasdeveloped to identif y each lot and the day of processing. Samples fromeach day' s pack were se sent to an independent labotatory and bacteriologicalanalyses were made te to count total bacteria, coliforurn bacteria andstaphylococci.

>. 4.4 Descri tion of 0 erations

Harvesting and processing equipment was assembled thx'o ghounovember and early December to conduct a trial run in mid Decembex'-Permission to begin harvesting the crop was received by the WashingtonState Department of Fisheries and a wholesale fish dealer's license w»obtained, The Department of Fisheries also ruled that the State privilegetax of 2'/o of the primary market value of all sales was applicable tofarm- rais ed salmon.

During December several txial runs were held in order to perfectthe harvesting process. In general, all worked well, although the znechanicalgrader ca.used considerable bruising and scale loss to larger fish. Inaddition, very active smaller fish often bypassed the grading slots insteadof dropping through. For these reasons, band grading was substituted and acrew of harvest workers replaced the mechanical grading systezn. Weatherconditions during winter znonths were sometimes severe and efficiencydropped. proper clothing under such conditions is vital and as much shelteras possible is important.

The harvest continued whenever fish reached market size until thetermination of the Pilot Project on June 30, l9'72. In addition to the fishwhich were harvested, others were released for the sports fishery,escaped, were released because they would not reach market size prioz-to the end of the Pilot Project, or were turned over to NMFS for reseazchprojects. In addition, approximately 5, ZOO were retained as broodstock for future selective bx ceding purposes. Unknown, but probablysignificant numbers of slow growing fish were lost to cannibalism.Frequentl smaQ salmon were found in the stomachs of larger fishrequen y, sma

being p ro ce s s ed.

-42-

The following table summarized the harvest:

Co

December 1971

January 197Z

February 197Z

March 197Z

A pril 1 97Z

May 197Z

June 197Z Total

Grand Total

Processing proceeded in a routine fashion as described earlier.These operations are shown in Figures 40 through 45 . Late in June,a small batch of 10-lZ -ounce salmon were boned in order to testcustomer reaction to the boned product.

3. 4. 5

In order to evaluate the market appeal of pan-size salmon, theDivision of Marketing Services of the National Marine Fisheries Serviceconducted a small-scale market survey during the spring of 1971. Samplesof the fish were sent to Z7 selected brokers and buyers throughout thenation. The response was generally favorable. A copy of the inquiry letteris provided in Appendix C.

After January 3, 197Z, pan-size salmon were introduced to theinstitutional seafood market by established seafood brokers. Fivemarketing objectives were given to the brokers:

a! Establish the pan-size salmon as a new high quality,gourmet item worthy of a premium price;

b! Concentrate on the prestige restaurant trade andestablish a repeat order business in a major city;

c! Establish an export market;d! Test the retail market;e! Introduce the fish selectively throughout the

United States in order to begin t' he developmentof a brood market base.

-43-

FrG! -R E 40

Butchering Operations, Fishare inspected and washed; thebellies slit and gills cut.Butchered fish are then placedon a slide for transfer to theeviserating station.

FIGURE 41

Eviscerating. Viscera andgills are removed and placed incontainers to be recycled intofish food. Eviscerated fish arewashed, trixzuned and inspectedbefore passed to the vacuumeviscerator to remove kidneys.

I iGl:R E 4Z

Kidney 8 ernoval. 1 he kidneyis sucked from the backbone

without scraping with the useof a vacuum e vis cerator. Thi s

approach serves to keep thebelly cavity and rib cage intactwhile thoroughly removingthe blood and kidney mater ial.

FIGURE 43

Inspection and Grading. Eachfish is weighed and separatedinto size categories. Gradedfish are stored in chilled

containers until each group isto be packed.

-45-

Packagtng Operations, Each fishis pLaced into an individual

polyethelene envelope. Thefish are then packaged into five-pound institutional cartons.

FICiURK 45

Boning. Small quantities of thefish were boned in order to conducttest marketing. The backbone wa,sremoved and rib bones strippedaway from the flesh.

-46-

All objectives were achieved and the product has been sold bywholesale distributors to restaur ants at approximately $l. 75 per pound.The retail market was tested on a small scale, but due to the shortage ofpzoduct and the high demand of the institutional market, the tesults werenot conclusive.

3. 5 Public Awareness

Since one of the primary goa' - of the Sea Grant Program was todi'sseminate information, an obje .ive of this project was to increasepublic awareness of salmon aquaculture. Simultaneously an informala.ttempt was made to assess public acceptance of salznon aquaculture a.san industry in the Puget Sound a,rea.

A variety of semi-popular articles were published in trade magazinesand newspapers, A partial list of these articles is included in Appendix A.In addition, a documentary film a,ppeared on Seattle television in fourinstallments during December 1971.

An extensive number of visitors toured the facility, including personsfrozn R ussia, Gzes t Britain, Argentina, Nigeria, Norway, Japan, Israel,Canada, South Africa, France, U. S. Senators and Congressmen,representatives from U. S. Government agencies, and state and localagencies. Public addresses reached such groups as the Port OrchardRotary Club, the U. S. Navy Technical Supervisors at Bangor, Washington,the County Planners Association held at Aberdeen, Washington, and variousaudiences through the country. A more coznplete listing is shown inAppendix 8,

In conjunction with local sportsmen groups and the WashingtonState Department of Fisheries, 631 market-size coho were tagged andreleased in January 1972. This effort was part of the In-So.und SalznonEnhancement Program which appears to indicate that salmon which arecultivated in saltwater pens for a significant time prior to release aremore likely to remain in Puget Sound rather than znigrate to the openocean. Releases of the tagged fish were znade at Manchester and in CaseInlet in southern Puget Sound.

During January, the Departznent of Fisheries conducted hearings inOiyznpia to assist in the developznent of mariculture procedu.res.Washington State legislation previously author ized the Department ofFisheries to establish such procedures and authorize commercial salznonaquaculture. In general, public comment was favorable, and the Departmentof Fisheries decided to proceed with the licensing of coznmercialoperations.

-47-

3. 6 Environmental iklonitorin Pro ram

The presence of large numbers of animals in a restricted spacemakes the problem of waste materials inevitable. As in any feed lotoperation, the salmon aquaculture project has the potential for contributingquantities of unused feed and fish waste material to the environment.In order to measure the direct effect of aquaculture in Clam Bay, weattempted to introduce a znonitoring program in the fall of 1971. Thisprogram was not part of the original plan, but was felt worthwhile.

It was decided that direct changes in water quality would be slightand difficult to interpret due to the high tidal exchange. The approach wasto measure changes in the degree and composition of sediment on the bottomin the vicinity of the pens, Changes in organic content and sedixnentationrate were considered to be the best indicators of the fish ws,ste influence.Four parazneters were to be checked:

a! Sediznentation rateb! Total organic carbonc! Total organic nitrogend! Quantitative changes in benthic fauna

A twelve-point 400 x 500 ft. grid ws.s established on the bottoxn bydivers, with the net pens at its center. One additional point was locatedat a distance of about 500 ft. from the grid. Depths of these saznplingsites range from 20 to 85 ft. at znean low water. A core sample was tobe taken at each point, refrigerated, divided into sections, and analyzedfor total organic nitrogen and tots.l organic carbon. Ballinger and hicKee�971! have shown that the percentage of organic nitrogen and carbon give agood cor relation with general pollutant loading and flow characteristics ofoverlying water bodies. The high organic content of fish wastes contrastswith the rela.tively low organic content of a sand and silt bottom.

When the cores were collected, dissolved oxygen readings were to betaken at the sazne location on the bottom and one xneter above the bottom,both at strong ebb and at near slack water' conditions.

Sedimentation was to be xneasured by installing cup- shaped containersat eight grid points. The collected material was to be collected and~eighed.

3. 6. 1 Environmental Monitorin 0 erations

In practice, inclement weather as well as operational problemscoxnproxnised the program to the point that little technical data was obtained.However, visual inspection in the vicinity of the floating net pens wasconducted twice each week.

-48-

Sediment consisting primarily of fecal material biit also of excessfood did biiild»p, partic»larly during the month before harvest. The de;ii'..of sediment va ried at its peak from three or four centimeters directlyunder the pens to a trace 100 feet peripherally i» the directions ofrz>axirnurn c»rre»t flow. Vo sediment was ol>served beyond l00 feetfrori~ the pens.

Most benthic fa«na and flora were destroyed in the L00 foot srpizruarea directly under the pens, probably due to high lLZS c»ncentr.~tin»In'a periphical zone approximately 30 feet siir rounding the pens, hentb«;organisirns were visibly affected. Beyond 30 feet from the pens, novisible deterioration was evident,

The National Marine E'isheries Services is conducting a detailedenvironmental study of the Pilot Project site beginning during the latterstages of fish rearing and continuing as the bottom reverts to normaL.The results of these studies will be published at a later date.

-49-

4. J CONCLUSIONS, DISCUSSION AND R ECO51MENDATIONS

The basic objective of the Pacific Aquaciilture Program was successful.Large numbers of coho salmon were cultivated at high concentrationsin floating net pens in the salt waters of Puget Sound. Growth rates andconversion rates ~ere good. The bulk of the fish r'cached market size atthree-quarter pound after fourteen nionths from spawning. The fishwere healthy and market acceptance seems good, We consider salmonaquaculture to be ready for full cornniercial development, albeit with substantialrieks. Significant problems do remain to be solved.

l, Adequate inventory control was not accomplished. Withoutexception, there were considerably less fish than expectedfrom proj ected write dowris and mortality counts. Thevariability was not constant. On the other hand, Petersonestimates - a sample of fish counted, tagged, and reLeasedto be statistically analyzed on an immediate subsequentsample - gave quite low estimates.

Loss of inventory control obviously means that businessplanning is difficult. It also means that control of feedingis lost. Feed cost and good conversions absolutely requireaccurate knowledge of the numbers and size of the group offish to be fed.

Beyond that, it is vital to learn about unexpected decreasesof fish quickly in order to determine the causes and taketimely action.

To maintain inventory control it is necessary to determinehow many fish are in a peri at a given time. Possibleanswers may be:

a! Seine, grade, and weigh the fLsh more frequently.However, this i ~ time consuming and stresses the fish.

b! Better ways of conducting Peterson estimates.c! Develop an insitu counter, such as an ultra sound

transponder system.d! Better control of the causes of losses and devising

means of detecting the existence of losses.

Unaccounted for losses of fish appeared. to come from severalsources:

a! iVets developed holes and rips. Heavier and betterdesigned nets, protection systems, and mooring systemsmu s t be de v eloped.

b! Cannibalism occurs when the larger fish are morethan three times heavier than the sznaller fish.

Sufficient grading will he necessary to prevent sucha wide s ize discrepancy.

c! Pzedatory sea birds were a chronic problezn. Adequatemeans to deter such birds must be developed.

d} Dogfish caused unknown escapes during the ear ly partof the experiment. Concentric nets providing aseparation were a successful solution.

e! Small fish disintegr'ate quickly and may have beenmissed by morta,lity counts.

f! Theft is an ever present concern as the fish reachmarket size. Thieves were chased off after beingseen trying to fish in the pens. Two lures werefound in the mouths of brood f'ish.

Z. The chinook did not do well. It is our opinion that the speciesis not suitable for coznmercial aquaculture at the present stateof the art.

a! They were considerably more subject to Vibrio at almostall saltwater stages of the project.

b! They did not adapt well to dry food in the freshwater stage.In salt water they did grow well with dry food in warmerwater, but grew better in colder water' with znoist food.

c! The chinook sustained far more losses frozn handlingand scale loss i.n salt water than coho.

d} Growth virtually stopped in October for the chinook,and they had to be carried through the winter.

e} The costs of raising a pound of chinook to znarket sizewere far higher than a pound of coho.

3. Significant nuznbers of the second group of coho, transferred tosalt water in August, did not begin to grow until spring. Thereare strong indications that coho must be transferred early inthe suznmer to be successful.

4, Underdesigned saltwater facilities caused serious maintenanceand logistic problezns. Anchors dragged, nets ripped, andcables and lines broke.

5. While present available foods are adequate, they are obviouslynot ideal. Further research is needed.

6. The fish raised in this experiznent were spawned frozn wildparents which were evolved frozn the necessities of lifein the sea. A fish best adapted for cultivation by man inthe environment of a. net pen could have znarked advanta.gesover a wild fish. Genetic selection of suitable strains forcultivation may well be the key to future aquaculture.

-5l.-

A PPENDfx

A, List of published newspaper and rnagasine reports.

B, List of presentations.

C. Market Survey inquiry letter sent by the National MarineFisheries Service, Division of Marketing Services.

APPENDIX A

LIST OF ARTICLES AND STORIES

Future

January issue - Sea Grant 70January issue - Oceanology InternationalDecember issue - Fish Farrnirg Industries

Fish Farmin Industries, October 1971, Vol. Z No. 4 p. 4 - Descriptionof project and feature article to appear in December issue.

IA ence France Pr esse, May 24, 1971

News Release, Office of Floyd V. Hicks and Julia Buther Hanserr,July ZO, 1971, describing serving of fish in Congressional diningrooms. Story was distributed to Seattle Post Intelligence,radio KCMC and KVI, KTAC, KTNT, Tacoma News Tribune,TV KING, KIMO, Brernerton Sun, Radio KBRO, Wa.shington Post.

Tacoma News Tribune, Malcom MacNey, series of six articles on aqua-culture in Puget Sound. Placed in Congressional Record onSeptember 9, 1971 by Representative Floyd V. Hicks, pp E9348-9352.

Seattle Post Intelli ence, Thursday, April ZZ, 1971. Feature articleentitled "Salmon Farm under Experiment in Sound", by Judy Isrnach.

The Seattle Times, January? 0, 1971. Feature article by Hill Williamsentitled "Sound Seen as Home of Salznon Farms".

The Seattle Times, October 17, 1971. Feature article by Hill Williamsentitled "Puget Sound Site of Salmon Farm Tests",

Seattle Post Intelli encier no daCe!. Article by Dan Page entitled"Salmon Project Funded".

The Bremerton Sun, March Z3, 1971. Article by Elling Simonsenentitled "Few Weeds Gro~ing in Ocean Farming".

The Bremerton Sun, July 6, 1971. Article by Elling Sirnonsenentitled 'Manchester Project Very Kncouraging",

Tacoma News Tribune, July Zl, 1971. Article by Malcom MacNeyentitled "Aquacultured Fish on Varied Culinary Diet".

A-I

The Brewer ton Sun, December I, 1970. Feature article by TravisBaker entitled "Optixnum Salmon - an Industrial $$$ Test atManchester".

The Seattle Times, December ZO, 1970. News story entitled 'UnderwaterFarms could be big industry, says Bert Cole Commissioner ofNational Resources!".

Parade Ma azine, May 9, 1971. Feature story entitled "Can FishFarms SoLve the Food Problem' ?" By Sid Ross and HerbertKupferberg, pp 12-13.

Commerce Toda June l4, l 97k, p, Z9. "Private Salmon VentureBoosted".

Tacoma News Tribune and Sunda Led er. Article entitled "Fish FarmersBreed Super Salmon with Diet Control".

Port Orchard Inde endent, Septexxxber 30, ! 970. "Aquaculture Project'may Lead to Industry of the Future for Kitsap County". Paper alsofeatured an editorial supporting the project.

Bainbrid e Review, January 2, 1971. "Raising Salxnon in Pens".

Bremerton Sun no date!. "Power Plants and Fish Farms".

Bremerton Sun, March ZZ, L971. Feature article by Elling Sixxxonsenentitled "Fish Farming Prospects Look Good".

Bellin ham Hea.rid no date!. "Sea Farming Experiment Appears to bes. Success".

Tacoxna New s Tribune, June 20, 1971. Editorial entitled "FarmingPug e t Sound" .

A-2

A PPEND IX B

PRESENTATIONS

1, Meeting of Rural Development Subcommittee of the Overall EconomicDevelopment Committee appointed by the Board of Kitsap CountyCommi s sioner s. October S, 1970. A s reported in the OverallEconomic Development Flan OEDP! for Kitsap County, Washington.November 1970.

?, Meeting of Natural Resources and Environment Subcommittee of theOverall Economic Development Committee appointed by the Board.of Kitsap County Commissioners. October 19, 1970, As reportedin the Overall Economic Development Plan OEDP! for KitsapCounty, Washington. November 1970.

3. Presentation of fish to the members of Congress by RepresentativesEloyd V. Hicks and Julia Butler Hansen.

APPENDS C Xationa.l Marine Fisheries ervice

Division of Market>ng Services2725 Montlake Boulevard East

March 15, 1971

SALTViATER FARMED SALMON

Historically, salmon have been an important food resource in the PacificNorth~est, To supplement natural production, various governmental agenciesoperate freshwater hatcheries where young salmon are reared until theyare ready to migrate downstream to the sea.

Rearing salmon to maturity in salt water would extend control over the entirelife cycle of these fish, thereby enhancing the capability for breedingselectively f' or characteristics hest suited to market demands. Researchof this nature is currently being conducted at the National Marine FisheriesService Aquacultural Experixnent Station located at Manchester, Washington,in Puget Sound.

The floating fish husbandry units consist of sxnall, wood-fra.xned nurserycages with nylon screens and larger pens made of weighted knotless nylonnetting, suspended between walkways supported on styrofoaxn floats.

Newly hatched fry are placed in circular tanks of fiberglass or of steellined with polyethylene sheeting. The tanks are supplied with fresh and saltwater. Salinity is adjusted by regula. ting the flow of each.

When transferred to the floating pens, the fish are fed with Oregon MoistPellets, a wet, high-protein feed, originally compounded for salxnonhatcheries. This feed is supplemented by the naturally-occurring planktonand other small forms of sea life that are carried in with the tidal currents.Growth under such ideal conditions is accelerated, considerably over thatnormally achieved in the natural pattern of extended freshwater life.Salmon reared by this method are ready for marketing as trout-sized fishin 18 months ox' less.

Saxnples of these saltwater-reared salmon are being distributed by theNational Marine Fisheries Service in an effort to obtain information froxnretailers, wholesalers and restaurant operators as to their interest inmerchandising them. Information is needed as follows:

�! Market form desired--dressed, head-on; dressed, head-off; or boned.�! Size or sizes preferred�! A cceptability of price�! Interest in merchandising these salmon

In January, 1972, a commercial enterprise which has recently entered thisfield, will have 400, 000 saltwater-reared salmon available to market. Thefish are being raised with technical assistance from the National MarineFisheries Service and will be available fresh and frozen throughout theyear.

C-1

If this initial sampling is successful, the company will raise approximatelytwo million fish to be marketed starting January 1973. Since the eggsfor these salmon must be hatched in August 197l, the information on mar ketpotential is needed prior to August.

Although it is difficult to predict at this time, it is expected that thewholesale price of the fish f. o. b. Seattle will be between $, 90 and $1. l $a pound, depending upon size and market form.

C-2