Shortbelly Rockfish, Sebastes jordan;:A Large Unfished Resource inWaters Off California

WILLIAM H. LENARZ

Introduction

The shortbelly rockfish, Sebastesjordani, is one of the more distinctivemembers of the 57 species of rockfish(genus Sebastes) which are reportedfrom California waters. The species ob­tains its common name from the factthat its vent is about midway betweenthe origin of the anal fi n and insertion ofthe pelvic fin, while the vent of otherspecies of rockfish is located fartherback, much closer to the origin of theanal fin. The shortbelly rockfish alsohas a more strongly incised tail and amore fusiform shape than mostrockfish.

Moser et al. (1977) noted that larvaeof shortbelly rockfish are the longest atbirth of eastern Pacific rockfish studiedto date. Also, the larval period is longprior to transformation to the juvenilestage when compared with other east­ern Pacific rockfish. These features, inaddition to the pigmentation pattern andmorphometrics of young larvae, areremarkably similar to those of theredfish or ocean perch, S. marinus,group of the Atlantic Ocean.

Another distinction of adultshortbelly rockfish is that it occurs inmidwater and away from underwater

William H. Lenarz is with the Tiburon Labora­tory, Southwest Fisheries Center, NationalMarine Fisheries Service, NOAA, 3150 ParadiseDrive, Tiburon, CA 94920.

ABSTRACT-Shortbellyrockfish, Sebas­tes jordani, appear to be abundant inCalifornia waters and present the potentialfor development ofa new large fishery in thearea. This paper contains: 1) A descriptionof the biology ofthe species, 2) first approx-

34

objects such as reefs or kelp more oftenthan most, if not all, Cal ifornia rep­resentatives of the genus. Occasionalcatches have been made by purse sein­ers fishing in southern California wa­ters, and the species predominated inmidwater and demersal trawl catches ofrockfish off central California in the1977 rockfish survey (Gunderson andSample, 1980).

Shortbelly rockfish have been re­ported from San Benito Island, BajaCalifornia, Mexico (Moser et aI., 1977)to La Perouse Bank, British Columbia(Miller and Lea, 1972). Miller and Leaalso report maximum depth as 283 m(155 fathoms) and maximum totallength as 305 mm (12 inches).

While there is no fishery forshortbelly rockfish, the species appearsto be very abundant in California wa­ters. Moser et al. (1977) estimated thatlarvae of shortbelly rockfish composed20 percent of all rockfish larvae taken ina sampling program off southernCalifornia and 12 percent off centralCalifornia in 1966. While the catch ofshortbelly larvae compared with allrockfish larvae was higher off southernCalifornia than central California, thecatch per standard haul was higher offcentral California (4.22 larvae) thansouthern California (2.65 larvae). Thebiomass in a limited area of its range,between Pt. Ano Nuevo and Pt. SanPedro, was estimated to be 295,000metric tons (t) from the results of the

imations of the effect offishing on the stock,3) a review of the rockfish survey resultswith regard to fishing, 4) a review of thepotential for development of a fishery, and5) a discussion of management options forthe fishery.

hydroacoustic and mid water trawlsegment of the rockfish survey) Ma­son l

).

Preparation of this paper wasprompted by the interest shown by bothfishermen and processors in developinga fishery for shortbelly rockfish. Theobjectives of the paper are to sum­marize information on the species and:I) Describe the biology of the species,2) make first approximations of thereaction of the stock(s) to various levelsof fishing, 3) review the rockfish surveyresults with regard to fishing, 4) reviewpotential of a fishery, and 5) discussmanagement options for the fishery.

Biology

Larval and Juvenile Stages

Moser et al. (1977) summarizedavailable information on larval andjuvenile stages of shortbelly rockfish.The larvae are released at an averagesize of 5.4 mm by the ovoviviparousfemales between January and April.Larvae were collected between north­ern Baja California, Mexico, and as farnorth as San Francisco, Calif. Theirsamples were not taken north of SanFrancisco. While larvae were collectedas far as 278 km (150 miles) offshore,all above-average catches were takenmuch closer to shore. They also re­ported that larvae metamorphose tojuveniles at 27 mm and appear to beginforming schools at the surface at thattime. Juveniles up to 62.8 mm havebeen taken by dip nets under nightlights. On the other hand, specimens assmall as 70 mm were taken by bottomtrawling during the rockfish survey.

Growth

Shortbelly rockfish, while being oneof the smaller species of rockfish, hasthe highest rate of growth completion(k) of 10 California species studied byPhillips (1964). The von Bertalanffy

'Mason, 1. E. 1978. Preliminary report on thehydroacoustic/midwater trawl survey for rockfishconducted off parts of the American and Cana­dian west coasts during July 12-September 30,1977. Unpubl. manuscr., 6 p. Northwest andAlaska Fisheries Center, NMFS, NOAA, 2725Montlake Blvd. E., Seattle, WA 98112.

Marine Fisheries Review

<l 5 7 8 9 10

28

fish from several sources and evalua­tion of selectivity is not possible. Datafrom the surveys (Fig. 2) indicated thatfemales grow larger than males and at­tain maximum size at a slower rate ofcompletion (lower k). Growth does notappear to be related to depth of capture,but there are insufficient data from thesurvey to arrive at conclusive results.

Length-Weight

The length-weight relationship wasestimated by Phillips to be log W =a +{3logL, where W = weight inpounds, L = total length (mm), a =-8.05202, and {3 = 3.1518.

Maturation, Fecundity, andSex Composition

Phillips (1964) reported that 50 per­cent of shortbelly rockfish " . . . aremature when 6Jh inches [16.5 cm]long, or 3 years old." The fecundity­length relationship, estimated by mefrom Phillips' data on to specimens isIn F = a + /3lnL, where F = fecundity(numbers of eggs), L = total length(mm), a = -8.43523, and f3 =3.30611.

About 45 percent of the surveycatches of shortbelly rockfish between14 cm and 27 cm were males. Few ofthe fish larger than 27 cm were males.

Movements

While many aspects of the life his­tory of shortbelly rockfish appear in theliterature, nothing could be found onmovements. Tagging studies on bluerockfish, S. mystinus, Miller andGeibel (1973); yellowtail rockfish, S.flavidus, Carlson and Haight (1972);copper rockfish, S. caurinus, Dewees(1970); and black-and-yellow rockfish,S. chrysomelas, and gopher rockfish,S. carnatus, Hallacher (1977) and Lar­son (1977) indicated little movement.However, these five species do not havestrongly incised tails or fusiformbodies. These aspects of morphologywould suggest that shortbelly rockfishare stronger swimmers and are betteradapted to swim away from predatorsand/or make more extensive move­ments than the other fi ve species (Hob­son and Chess, 1978). Analyses ofcatch rates of Pacific ocean perch, S.

Moles

Age (yeors)

Figure 2.-Von Bertalanffy growthcurves for female and male Sebastesjordani. L x is 324 mm for femalesand 290 mm for males; k is 0.2112for females and 0.2980 for males.

Age (yeors)

1 2 3 " 5 6 7 8 9 10

20

24

Females

Figure I.-Von Bertalanffy growthcurves for Sebastesjordani estimatedfrom this study and by Phillips(1964).

32

2.

2'

20

~.c

'" I.e~

;;

~ 12

mesh size of the nets used by the surveymay have selected for relatively largefish at young ages. Phillips obtained

Phillips combined age-length data fromboth sexes and made his estimates fromback-calculated lengths (estimates oflength at time of formation of each an­nual ring on a scale) made from scalereadings and measurements. He as­sumed that scales form at birth. Myexperience with several other species ofrockfish indicates that scales form whenthe fish reach about 20 mm. Thus theback-calculations of Phillips probablyslightly under-estimate lengths ofyoung fish.

Additional data on growth are avail­able from the rockfish survey. Otolithsfrom 1,081 specimens were read.Opaque zones of fast growth were justbeginning to be formed at the time ofsampling (midsummer). Phillips'back-calculations appear to be at theend of the growth season (midwinter).Thus fish from the rockfish surveyshould be 0.5 year older than those Phil­lips used. Phillips used total lengths,while fork lengths were measured forthe rockfish survey. Total length isabout 15 mm greater than fork lengthfor shortbelly rockfish, and the surveydata were adjusted accordingly. Theprogram BGC2 (Abramson, 1971) wasused to estimate parameters of thegrowth curve. The estimate of k for thesurvey data, 0.27721, is very close tothe estimate by Phillips, 0.27520. Theestimates of L x are also close, 315(Phillips) and 30 I (survey). Compari­son of the two growth curves (Fig. I)reveals that the main differences occurat young ages. As previously men­tioned, Phillips probably underesti­mated length offish at young ages. Also

growth curve was used to describegrowth:

Lf

= L x(l-e -k(f-fO»)

where L f = total length (mm) att,

t = age in years,k = growth completion

rate,to= theoretical age when

fish is length 0,and

Lx = estimate of averagelength attained atmaximum age.

March-April 1980 35

Figure 3 .-Length composition of Sebastes jordani caught by bot­tom trawling during rockfish survey as a function of depth andlatitude.

'::UL C C~228mLN.44'59'N

WOOLLLrr

~N'39'59'N5000

4000L LfuL LN.3B'59'N20000

£

'" 2o.000L'0 10.000 lOLL

LillN.37' 59'N:u.0 0E:>Z 40.00°

lnL20,000LL

LN'36'59'N0

400°bnLLLLN.35'59'N20000

2000UlLL~

LN'34'59'N10000'0 20 30 10 20 30 10 20 30 10 20 30

Fork length (em)

35° 4S'N95m

36° 59'N114m

37° lS'N144m

'~~I 2 --:3=--,.,---'-75 -'-:',-'-:'-'-:.,..-'-9:::-'--,""0-"--:";-'''

Age (yeors)

,OOO~37" S'N

200: 176m

JO.OO°lrll20,000

10,000

o '----~-~--~~

2o'OOOh10.000

o ~-~~-~~~

Figure 5.-Age composition of bot­tom trawl catches ofSebastes jordaniduring rockfish survey.

found that rockfish represented 22.5percent of the food volume of chinook

~J9'N-39"S9'N

o 44° N·44° !)9"N

~r ~v- L-L-J 38"N·JS'S9'N

Fork lengfh (em)

10.~[ I I h37'N.370S9·N

2O.~[ c=O 36° N-36°S9'N

~[ c=[]~~

~~"N10 15 20 ~ ~ ~

Figure 4.-Length composition ofSebastes jordani caught by bollomtrawling during rockfish survey as afunction of latitude.

Relationships With Other Species

Shortbelly rockfish appear to be animportant prey item in central andsouthern California. Merkel (1957)

alutus, indicate considerable move­ment into relatively shallow waters dur­ing spring and summer and return todeeper water in the winter (Gunderson,1972). Pacific ocean perch have morestrongly indented tails and fusiformbodies than the five species that weretagged but less than shortbelly rockfish.Sorokin (1961) documented long (atleast I ,112-krn (600-mile» migrationsof stocks of redfish (or ocean perch),and deep- water redfi sh, S. mentella, inthe Atlantic Ocean.

Data from bottom trawl catches ofthe rockfish survey indicate thatshortbelly rockfish do make significantmovements. Length and age sampleswere expanded to entire catches assum­ing that sampled fish were taken at ran­dom. Size of fish tends to increase withdepth (Fig. 3). Size for a given depthstratum tends to increase in a northerlydirection between lat. 34° and 39° N.Depth of capture of shortbelly rockfishalso tends to increase in a northerlydirection between lat. 35° and 39° N.The combined results of these trends isa very strong tendency for size to in­crease in a northerly direction betweenlat. 34° and 39°N (Fig. 4). While agecomposition data are not sufficient toseparate effects of depth and latitude,age of shortbelly rockfish tends to in­crease as depth and latitude increase(Fig. 5). The age and length composi­tion data from the rockfish survey indi­cate that at least during summer,shortbelly rockfish tend to move intodeeper waters and to the north as theygrow.

Data on larvae revealed areas ofheavy spawning from Los Angeles to atleast San Francisco during January1966 and March 1975. 2 Very heavyspawning occurred off Monterey (lat.37°N), Avila Beach (lat. 35°N), andSanta Barbara (lat. 34ON). Heaviestspawning occurred off Monterey. Theheavy spawning from Los Angeles to atleast San Francisco in the same monthindicated that if shortbelly rockfish dotend to move north as they grow, theydo not make long return migrations tothe south in the winter to spawn.

2Unpublished data of H. G. Moser, SouthwestFisheries Center, NMFS, NOAA, P.O. Box 271,La Jolla, CA 92038. Pers. commun. 1978.

36 Marine Fisheries Review

Figure 7. - Relationshi ps between length oflongest gill raker and fork length (a) and lengthof upper jaw and fork length (b) for Sebastesjordani, S. goodei, and S. paucispinis.

::l Sebosles jordoni L, , , , , , , , , , ! , ,o '

25

20.c

Sebosles goodeiV>'5

'0 '"~

-rrf.0 5

~ 0 lz

20

15

Sebasles poucispinis

'" rh',r0

26 26 30 32 34 36 38 '0 " " " " 50

Number of gill rakers

Figure 6.-Gill raker frequencies ofSebastes jordani, S. goodei, and S.paucispinis.

20 20

16 16E.s~ §E 12 112'0.

~

i 0.

~~ '0"0 8 .c 8.c '"c1> ~

C ...J

~

a b0

20 40 60 20 40 60

Fork length (em)

salmon, Oncorhynchus tshawytscha,off San Francisco between Februaryand November. In June and July morethan 50 percent of the food wasrockfish, of which 70 percent wasshortbelly rockfish. Few of the rockfishexceeded 89 mm (3.5 inches). Short­belly rockfish less than 89 mID areprobably in their first year of life. Of 78fish sampled from pigeon guillemots,Cephus columbia, nesting on the Faral­Ion Islands, two were shortbellyrockfish that were 59 and 83 mm instandard length (Follet and Ainley,1976). Hubbs et al. (1970) reported that4 otoliths of shortbelly rockfish werefound in 498 otoliths sampled fromwestern gull, Larus occidentalis, cast­ings near San Diego. The fish werethought to have been stolen fromBrandt's cormorant, Phalacrocoraxpenicillatus. Trawl-caught bocaccio,S. paucispinis, and chilipepper, S.goodei, are often seen with shortbellyrockfish protruding from their mouths,but these specimens may represent un­natural feeding within the trawl. Anumber of studies on food of fish,marine mammals, and birds foundsmall rockfish to be important preyitems, but did not identify the items tospecies. In summary, at timesshortbelly rockfish can be an importantprey item for chinook salmon and pos­sibly other fish. They also compose yet

March-April 1980

undetermined portions of the diets ofsome birds. Only two studies reportedthe size of consumed shortbellyrockfish. In both cases small fish, prob­ably less than 1 year old, predominated.

Shortbelly rockfish belong to a groupof important rockfish that inhabitCalifornia waters between 50 and 125fathoms (91 and 229 m) during adultlife. The other two members are bocac­cio and chilipepper, both of which arevery important components of Califor­nia rockfish landings. All three speciesare relatively fusiform and have deeplyincised tails for rockfish.

Since the three species are oftencaught together and belong to the samegenus, ecological theory requires thatthere be mechanisms that separate thespecies into different niches. Mor­phological features, measured follow­ing procedures of Phillips (1957), re­lated to feeding differ considerablyamong the species. For example, thereis almost no overlap in gill raker countsfor the three species (Fig. 6). However,the relationship between length of thelargest gill raker of the first arch andfork length is similar for chilipepperand shortbelly rockfish, but differs be­tween these two species and bocaccio(Fig. 7a). There are considerable dif­ferences in the relative length of upperjaw among the species (Fig. 7b). Thedata presented in Figures 6 and 7

suggest that shortbelly rockfish con­sume the smallest organisms, probablyplankton, of the three species andbocaccio consume the largest, probablyfish. Phillips (1964) stated that theshortbelly rockfish " ... feed exclu­sively on macroplanktonic organisms,primarily euphausids," while chilipep­per" ... may feed on euphausids andsmall squids, or on small fishes such asanchovies, young hake, Merlucciusproductus, and lanternfish," andbocaccio " ... feed on smallerrockfishes, sablefish, Anoplopomafimbria, anchovies, lantern-fish, andsquids, Loligo opalescens."

A compilation of upper jaw lengthfrequencies of the three species inrockfish survey tows containingshortbelly rockfish indicates little over­lap (Fig. 8). Since young specimens ofthe larger species have jaw lengths ofthe same size as older specimens of thesmaller species, the lack of overlapshown in Figure 8 suggests that thethree species are distributed in a fashionthat minimizes the possibility of com­petition for food.

First Approximation ofReaction of Species to Fishing

Yield per recruit calculations oftenare used to make first approximations ofproduction of a stock under differentfishing strategies. Estimates of growth

37

04

0.40.3

06 05

0,20.1

b0,4

Fishing mortality

0.30.20.1

a

~Sebasles gaadei

:2'""0~

JJ:JE~

z

'"0--'

Sebasles pauclsplnis

Length of upper jOW(CM)

Figure 9.-Yield per recruit (a) and relative fecundity (b)(production of larvae by fished stock/production of lar­vae by unfished stock) of Sebastes jordani as functions ofage at recruitment and instantaneous rate of fishing mor­tality (F).

Figure 8.-Frequencies of upper jawlength of Sebastes jordani. S.goodei. and S. paucispinis caught inrockfish survey bottom trawls con­taining S. jordani.

and mortality are needed for the calcu­lations. Growth estimates have alreadybeen presented. No estimates of mortal­ity rates were found in the literature.Often the rate of natural mortality (M)is approximately equal to k (Bevertonand Holt, 1959). Thus a first approxi­mation of M is 0.275. Of 1,081shortbelly rockfish aged from rockfishsurvey catches the oldest were 12 yearsold, which is older than the lO-yearmaximum age reported by Phillips(1964). It is likely that, if M is sig­nificantly lower than 0.275, older fishwould occur in the samples. Indeed, ahigher value of M would be reasonable.

Assuming that natural mortality isconstant and that fishing mortality isconstant beyond age at recruitment, Iused the Ricker (1958) yield equation toestimate yield per recruit at three agesof recruitment: 2, 3, and 4 years. Re­sults of the calculations indicate thatyield per recruit is not sensitive to ageof recruitment at F (fishing mortality)values less than 0.25 (Fig. 9a). When Fis greater than 0,25, yield per recruit forrecruitment at ages 3 or 4 is greater thanwhen age at recruitment is 2 years.There is little difference in yield perrecruit when age at recruitment is 3 or4for the values of F between 0,05 and0.4.

The fecundity of shortbelly rockfishis similar to the clupeids which have awell-documented history of recruit­ment failures in heavily fished stocks(Murphy, 1977). With this in mind, theeffect of fishing on production of larvaeby a population of shortbelly rockfishwas estimated. Fish were assumed tospawn once at the beginning of theyear, the maturity schedule andfecundity-age relationship described inthe Biology section were used, andfishing rates were assumed to be equalfor males and females. The results (Fig.9b) indicated that fishing would have aconsiderable effect on relative fecun­dity (production of larvae of fishedstock/production of larvae of unfishedstocks), and that relative fecundity isvery sensitive to age at recruitment athigher levels of fishing mortality,

Review of Rockfish SurveyResults With Regard To Fishing

N umbers of catches of shortbellyrockfish exceeding 450 kg (1,000pounds) were made by O. 5-hour bottomand midwater trawls during the rockfishsurvey and a pilot survey (Gundersonand Nelson3 ) in the Monterey area dur-

3Gunderson, D., and M. Nelson. 1977. Prelimi­nary report on an experimental rockfish surveyconducted off Monterey, California. and inQueen Charlotte Sound, Blitish Columbia, dur­ing August-September 1976. Unpubl manuscr,82 p. Northwest and Alaska Fisheries Center.NMFS, NOAA, 2725 Montlake BI vd. E., Seat­tle, WA 98112.

ing 1976 (Table I). An area of highcatches by both gears in both years wasbetween lat. 36°56' and 3r21'N.Large catches were made by both gearsbetween lat. 36°18' and 36°30'N in1976 but not in 1977 even though ninebottom trawl hauls were made in thearea in 1977. Most large midwatercatches were taken by trawling close tothe bottom. The midwater trawl catchesaveraged higher than the bottom trawlcatches, 3,728 and 1,952 kg, respec­tively, probably because bottom trawlswere set at preselected stations, while

Table 1.-Catches of shortbelly rockfish. Sebastes jor-dani. that exceeded 450 kg in onl>-half hour bottom andmidwater trawls during 1977 rockfish survey and 1976pilot survey.

Position at begin- Depth (m)ning of tow

Bot- CatchDale La! (N) Long. (W) Gear tom (kg)

Bot/om trawls8-15-76 36'42' 121'58' 101 101 4598-19-76 36'20' 122'3' 212 212 6278-19·76 36'30' 122'0' 177 177 4,1207-18·77 35'33' 121'12' 132 132 5197-27-77 36'59' 122'22' 124 124 4.7227-27-77 37"10' 122'43' 176 176 8817-30-77 37'19' 122'50' 144 144 2,7907-30-77 37'18' 122'53' 240 240 2,4998-4·77 38'21 ' 123'25' 144 144 954

Midwater trawls8-12-76 36"56' 122'20' 154 229 3.3148-13-76 37'11' 123'48' 302 302 3,1568-13-76 37'12' 122'50' 282 285 1,4158-19-76 36'18' 122'3' 161 165 3,3138-20-76 36'21' 122'0' 274 274 7.2647·27- 77 37'13' 122'48' 245 247 4,0557·27-77 37'21' 122'52' 188 192 3,578

38 Marine Fisheries Review

midwater trawling was directed at con­centrations detected through echosounding.

Aimed midwater trawling couldprove to be an efficient method of har­vesting shortbelly rockfish which formlarge, off-bottom schools which arereadily detected by echo sounding. Thesurvey work indicated that with rela­tively little experience fishermen couldlearn to distinguish shortbelly rockfishfrom other abundant school ing speciessuch as Pacific whiting which is thespecies most likely to contribute to un­wanted by-catches.

Because of the small size ofshortbelly rockfish, trawls would haveto be constructed of mesh size smallerthan allowed for bottom trawls byCalifornia regulations, 4.5 inches(11.43 cm). Cod ends used in therockfish survey were constructed of1.25-inch (3.18 cm) webbing. Meshsize experiments are required to deter­mine optimum size for shortbellyrockfish.

The present California regulationsare designed to prevent capture offlatfish and other rockfish that aresmaller than desired. Thus by-catchesof other fish should be considered. Inthe belief that fishermen targeting onshortbelly rockfish would limit their ef­fort to areas producing large catches ofshortbelly rockfish, catch compositionsof hauls that produced more than 450 kgof shortbelly rockfish were compiledfor bottom and midwater trawls madeduring the rockfish and pilot surveys.

The catch by weight of species otherthan shortbelly rockfish was considera­bly higher for bottom trawling (23.67percent) than for midwater trawling(3.04 percent). Most of the by-catchwas Pacific whiting for midwater trawl­ing, while the by-catch for bottomtrawling was dominated by chilipepper;bocaccio; Pacific whiting; jack macker­el, Trachurus symmetricus; and spinydogfish, Squalus acanthias. Only 0.54percent of the mid water trawl catcheswere composed of target species ofdomestic fishermen: Chilipepper,bocaccio, and sablefish. A relativelyhigh 10.39 percent of the bottom trawl

March-April 1980

catches were composed of targetspecies: Chilipepper; bocaccio;darkblotched rockfish, S. crameri,Dover sole, Microstomus pacificus; rexsole, Glyptocephalus zachirus; andjack mackerel.

The by-catch in a shortbelly rockfishfishery evidently can be kept at lowlevels by using only midwater trawls,but it would also be desirable to controlage at recruitment. If mesh size restric­tions prove impractical in accomplish­ing this, the relationships between sizeand age composition with depth andarea of bottom trawl hauls indicatedthat fishermen may be able to minimizecatches of relatively small or youngshortbelly rockfish by choice of depthand area fished (Fig. 3,4). Similar rela­tionships are apparent between depth ofhaul and length compositions of largecatches of shortbelly rockfish made bymidwater trawls. Five out of nine largecatches were made at depths exceeding182 m (99.5 fathoms), and fish smallerthan 200 mm fork length were notcaught in significant numbers at thesedepths. Thus it appears that, at leastduring summer, fishermen may be ableto minimize catches of small shortbellyrockfish through choice of depth fished.If mesh size restrictions prove practi­cal, it appears that catches would bemaximized by fishing at relatively deepdepths.

Potential for Developmentof the Fishery

S. Kato, a fishery biologist specializ­ing in fisheries development at theTiburon Laboratory of the NMFSSouthwest Fisheries Center, has iden­tified several potential markets forshortbelly rockfish. There may be asmall export market for good qualitywhole frozen fish. Another possible useis canning for either pet or human con­sumption, and a small test pack foundthe product acceptable as pet food. Apossibly very large market exists foruse as a raw product for surimi. Theflesh appears to be of suitable qualityfor surimi, but their small size maymake shortbelly rockfish more expen­sive to process than species currently

used. Kato has tentatively estimatedthat the price to fishermen would beabout $100 to $200 per ton. Actualprices, of course, will depend on anumber of economic factors that canonly be evaluated by fishermen andprocessors, and negotiations betweenthe two parties.

High catch rates are necessary tosupport a fishery for a species that has aprice range between $100 and $200 perton. It would probably be necessary fora vessel to capture 10 to 20 tons per dayto exceed operating costs. In addition tooperating costs, capital outlays wouldbe necessary to equip bottom trawlersfor midwater trawling.

While the potential market, largebiomass, and large catches shown inTable I indicated that there is consider­able potential for development of afishery for shortbelly rockfish, assis­tance by government agencies may benecessary for fullfillment of the poten­tial. Kato has begun the developmentprocess by locating potential markets,purchasing appropriate fishing gear,and working with fishermen and pro­cessors in initiating a test fishery.

Management Options ForFishery Development

Management (State of California,Pacific Fishery Management Council,and U.S. Department of Commerce)has the option of either allowing or notallowing a fishery for shortbellyrockfish to develop. An argumentagainst development of a fishery is thatshortbelly rockfish have been shown tobe prey items for important speciessuch as chinook salmon (Merkel,1957). On the other hand, Merkel'sstudy showed that while shortbellyrockfish composed about 16 percent ofthe food of chinook salmon, 51 percentof the food was composed of northernanchovy, Engraulis mordax, Pacificherring, Clupea harengus pallasi, andsquid (mostly Loligo opalescens). Sig­nificant fisheries have existed on thesespecies without any documented evi­dence of causing feeding problems forchinook salmon. In addition Merkel es­timated that euphausids composed 15

39

percent of the food of chinook salmon.Since shortbelly rockfish mainly feedon euphausids (Phillips, 1964), theymay compete with chinook salmon.

Management also has the option ofplacing restrictions on types of gear,seasons, and locations. The results ofthis study suggest that it may be advis­able to allow the use of midwater trawlsbut not bottom trawls because of rela­tively large by-catches of species otherthan shortbelly rockfish when bottomtrawls are used. While yield per recruitand population fecundity analyses indi­cated that age at recruitment shouldprobably be 3 or 4 years if a significantfishery develops, information is notavailable to make specific recommen­dations on obtaining that age at re­cruitment. The data (available only forsummer months) indicated that fisher­men may be able to avoid capture ofsmall fish by choice of fishing depthsand locations. No data are available onthe practicality of using mesh size re­strictions as a means to obtain a desiredsize at recruitment. Mesh size experi­ments and test fishing are required todetermine the correct policy for obtain­ing a specific size at recruitment.

Estimates of a large biomass ofshortbelly rockfish (295,000 t in a smallportion of its range) indicated that thepopulation could support a largefishery. However, the relatively lowfecundity of the species suggests thatcaution in developing the fishery wouldbe prudent. A relatively small fishery,e.g., 5,000-10,000 t of fish, may bedesirable until more is known about thespecies. A small test fishery wouldallow development of procedures to ob­tain desired size at recruitment throughmesh size experiments and monitoringof size and age composition of the catchwith respect to depth and area of catch.Location and extent of areas containinghigh densities of fish would providemore information on size of thestock(s). Further analyses of larval sur­vey data may provide additional esti-

40

mates of stock size. Hydroacousticmidwater trawl surveys probably couldbe designed to produce estimates ofstock size in areas of high densities ofthe species at relatively low cost. If theabove data indicate the possibility ofmore than one important stock, appro­priate studies should be carried out todetermine the stock structure. Finally,additional work, such as predator-preystudies, on the position of shortbellyrockfish in the community is desirable.Ideally, management should take intoaccount the effect of fishing onshortbelly rockfish on both its predatorsand competitors. However, as is thecase for other marine fisheries, little isknown on the interactions of shortbellyrockfish with other species.

Acknowledgments

Much of the data used for this reportwas collected by the 1977 rockfish sur­vey and the 1976 pilot survey. The ef­forts of the skippers, crews, and scien­tific parties of the vessels John N.Cobb, Commando, Miller Freeman,David Starr Jordan, Pal San Marie,and Pacific Raider are appreciated.

Special thanks are due to D. R. Gun­derson and M. O. Nelson of the NMFSNorthwest and Alaska Fisheries Centerand H. G. Moser and S. Kato of theNMFS Southwest Fisheries Center forfreely supplying unpublished informa­tion and reviewing this paper.

Literature Cited

Abramson, N. J. 197 I Computer programs forfish stock assessment. FAO Fish. Tech. Pap.101, var. pag.

Beverton, R. 1. H., and S. J Holt. 1959. Areview of the life spans and mortality rates offish in nature, and their relation to growth andother physiological characteristics. In G E.W. Wolstenholme and M. O. Conner (editors),Ciba Foundation colloquia on ageing, Vol. 5.The lifespan of animals, p 142-177. Little,Brown, Boston.

Carlson, H. R., and R. E. Haight. 1972. Evi­dence for a home site and homing of adultyellowtail rockfish, Sebasresfiavidus J Fish.Res. Board Can. 29:1011-1014.

Dewees, C. M. 1970. Population dynamics andfishing success of an artificial reef in HumboldtBay, California. M.S. Thesis, Humboldt StateUniv., Arcata, Calif., 77 p.

Follet, W I., and D. G. Ainley. 1976. Fishescollected by pigeon guiliemOls, Cephus col­umba (Pallas), nesting on southeast FarallonIsland, California. Calif. Fish Game62:28-31

Gunderson, D. R. 1972. Evidence that Pacificocean perch (Sebasres alurus) in Queen Char­lotte Sound form aggregations that have differ·ent biological characteristics. J. Fish. Res.Board Can. 29: 1061 -1070.

Gunderson, DR., and T. M. Sample. 1980.Distribution and abundance of rockfish offWashington, Oregon, and California during1977. Mar. Fish. Rev. 42(3-4):2-16.

Hallacher, L. E. 1977. Patterns of space and fooduse by inshore rockfishes (Scorpaenidae:Sebasres) of Carmel Bay, California. Ph.D.Thesis, Univ. Calif., Berkeley, 115 p.

Hobson, E. S., and J R. Chess. 1978. Trophicrelationships among fishes and plankton in thelagoon at Enewetak Atoll, Marshall Islands.Fish. Bull., U. S. 76: 133-153.

Hubbs, C. L., A. L. Kelly, and C. Limbaugh.1970. Diversity in feeding by Brandt's cor­morant near San Diego. Calif. Fish Game56: I56- I65.

Larson, R. J. 1977. Habitat selection and territo­rial competition as the caw,es of bathymetricsegregation of sibling rockfishes (Sebasres).Ph.D. Thesis, Univ. Calif .. Santa Barbara,170 p.

Merkel, T. J 1957. Food habits of the king salm­on Onchorhynchus rshawylscha (Walbaum),in the vicinity of San Francisco, California.Calif. Fish Game 43:249-270.

Miller, D. 1., and 1. 1. Geibel. 1973. Summaryofblue rockfish and lingcod life histories: a reefecology study; and giant kelp, Macrocysrispyrifera, experiments in Monterey Bay,California. Calif. Dep. Fish Game, Fish Bull.158,137 p.

_____, and R. N. Lea. 1972. Guide to thecoaslalmarine fishesofCalifomia. Calif. Dep.Fish Game, Fish Bull. 157, 235 p.

Moser, H. G., E. H. Ahlstrom, and E. M.Sandknop. 1977. Guide to the identification ofscorpionfish larvae (family Scorpaenidae) inthe eastern Pacific with comparative notes onspecies of Sebasles and Helicolenus from otheroceans. U. S. Dep. Commer., NOAA Tech.Rep. NMFS Circ. 402, 7 I p.

Murphy, G. I. 1977. Clupeoids. In J. A. Gulland(editor), Fish population dynamics, p. 283­308. John Wiley & Sons, N.Y.

Phi lIips, J. B. 1957. A review of the rockfishes ofCalifornia (family Scorpaenidae). Calif. Dep.Fish Game, Fish Bull. 104, 158 p.

1964. Life history studies on tenspecies of rockfish, (genus Sebastodes). Calif.Dep. Fish Game, Fish Bull. 126,70 p.

Ricker, W. E. 1958. Handbook of computationsfor biological statistics of fish populations.Fish. Res. Board Can., Bull 119,300 p.

Sorokin, V P. 1961. The redfish; gametogenesisand migrations of the Sebasles marinus (L.)and Sebasles menrella Travin. Rapp. P.- V.Reun. Cons. Int. Explor Mer 150:245-250.

Marine Fisheries Review