lCES 1990 PAPER C.M. 1990/G:18Sess. 0

•

•

THH DAILY FOOD CONSUKPTION OF SIGANUS FUSCESCENS~

AN IKPORTANT HERBIVORE IN SEAGRASS COHKUNITIESAT BOLINAO. PANGASINAN PROVINCE. PHILIPPINES

by

Homer B. HernandezPorfirio H. Alino

Marine Science InstituteUniversity of the Philippines

Diliman, Quezon City, Philippines

and

Astrid JarreInternational Center for Living

Aquatic Resources Management (ICLARM)Me P.O. Box 1501, MakatiMetro Manila, Philippines

ABSTRACT

The feeding habits of the rabbitfish Siganus fuscescens(Pisces: Siganidae) were investigated in the reef flats ofBolinao, Pangasinan, Philippines. A total of 579 specimenswere collected from 1989 to 1990. The sampling was donequarterly at approximately bihourly intervals during a 24hours cycle. The species feeds exclusively during daytime.Examination of diet composition yielded seagrass as the mostabundant food item (50-70% by volume). Stomach contentweight analysis yielded an average daily ration of 9.46 gfor fish of a mean wet weight of 27.9 g, from whichpopulation-based consumption was estimated. About 6.4% ofthe seagrass leaf production of the study site are consumedby this species.

\ .

INTRODUCTION

c

Siganids or rabbitfishes (Pisces:' Siganidae) are among thc

most irnportant' fish groups in Bolinao, Pangasinan in terms of

their ecological as weIl as their cornrnercial value. Among the

several siganid species present in the area, the most dominant is

the white-spotted rabbitfish, Siganus fuscescens (sornetirnes

e,rroneously referred to as ~ canaliculatus or s. oramin).

tt ~lthough about 780 fish species have been r~ported for this area,

the fishery 1s highly dependent on S. fuscescens both for the

juvenile and adult stages (Aragones 1987, DeI Norte and Pauly•

I

1990), which are local.ly known as " padas" and "barangen",

respectively. §..:. fuscescens is the biggest single species caught

(Aragones 1987, Del Nortc and Pauly 1990), i.c., contributes,

around 40 % of the total catch of the fishermen (DeI Norte et al •.

1989). The adults are sold fresh or dricd while the juveniles are

made into fish paste.

A review of past. research jn siganids is available

(Lam, 1974), and important contribution's on siganid biology and

population dynamics published since include Nay. et 'al. (1974),

Bryan (1975), Woodland (1979), Mangaoang (1982), Aragones (1987),

de la Paz and Aragones (1987), and DeI Nortc ~nd PauIy (1990).

Being hardy and fast' grmling, siganids havealso been proposed

for mariculture (Ablan and Rosario 1961, ßen-Tuvia et al. 1973,

von \'1esternhagen 1973, Horstmann 1975, 'von Westernhagen and \I

Rosenthal 1975;, 1 ~76, B'tm thondi 1981,. Baga and Sacayanan 1980,

Luchavez and Carumbana 1982, Avila 1984).

1

, ,

The present study was performed in order to 1) determine the

feeding habit and diet compositiQn, 2) cstimatethe ingestion and

evacuation rates, and 3) compare the estimates of the daily food

ration of Siganus fuscescens during different periods of the

year.

MATERIAL AND ~ffiTHODS

Bolinao is situated atthenorthwest~rn part of thc

Philippine Archipelago,near the mouth of Lingayen Gulf,Luzon.

The study's'ite is part of the fringing reef along the north

eastern side of nearshore Santiago rsland on the lagoonal

seagrass flats, covering an area of 24 km 2 • This is a site of

regular gill net and spear fishing activity for ~ fuscescens.-

The. site is also known for the biannual catches in fish corrals

("baklad") of big sch,?ols of juvenile siganids.,

Fish sampling was done four times in the year at

approximately quarterly intervals in the periods July-August

1989, November 1989, February1990 and May 1990. A tot~l of 579

fish were collected at approximately every two hours during a 24

ho ur cycle. Each bihourly sample was subdivided into two

. subgroups, a) for the analysi~ of' diet composition and ~) fr6m~

the estimation of feeding ratesfrom thc stomach contcnt wcights

(see also Table 1 for details on the sampling, procedure). During

the day, samples were taken from gill nets ("tabar ll), whcrcas thc

fish \lere speared during nighttime. Different fishing methods

were applied due to"thefishermen's obseivations that the subject

2

..

I ~

\ .

speeies are inaetive at night and thus cannot be c~ught by gi11

nets. eare was taken to seleet only nighttime fish sampies with

guts undamaged by the spearing process. Thc samp1cs" were put" on

iee immediate1y, transported to thc laboratory and deep frozen

there.

From the thawed specimens, various morphometric va1ues were

taken prior to any disseetion. To determine possible differenees

relative to size in rates of ingestion and evacuation, size

e frequency analysis was" performed. This did not show any elear

bimodal distribution (Fig. 1), so the data were not separated

into different size classes. Hmvever, thc two largest f1sh from

theAugust-September sampies were ornitted from further analysis.

The fish \wre subsequent1y disseeted \'1 i thout damaging the

stomaeh, whieh was theri separated from thc fish. The stomach

content was removed, its v01ume obtained bythe water

disp1acement method (He11awelland Abe1 1971) and the wet weight

taken. Ta estimate the maximum stomaeh v01ume, the,empty stomaeh

was sealed by artery forceps at both ends and gently injeeted to

fullness with water using a gradu~ted syringe.

In the sampies for thc analysis of food composition,

contents were taken out and bathed with freshv/ater in a petri

dish~ The food was separated" into gross categories (e.g.

llcrustacea", " a l gae", etc.), the relative proportion "(by "olurne)"

of which was then estimated and quantified inan arbitrary scaleiI

of 0 to 5 where, 0= 0-1%, 1= 2-10%, 2= 11-25%, 3=,26-50%, 4= 51-

70%, and 5= 71-100%.

3

"'._----------- ,

The stomach contents ofthc [ish samplcd werc recomputed for

an, "av'crage fish" based' on tbe mean vlet weight of the total (24

hours) ,'sample and the values for e<lch ~ampling time were

<lveraged. Daily T<ltion was estim<lted <1ccorc1ing to Jarre et a1.

(1989) from a slight modification of two !l1odelscomputing thc

time ,t r a j e c tory 0 f' s tOmach co n t e n t s pr c sen:t e d b y Sa insbur y

(1986). The first ("Model 1") ,is based on a constant ingestion

rate; the second ("Hodel 2") is based on an ingestion ,rate

e decrcasing with time, i.c., with 'increasing stomach contents,. Für

both models," simple exponential evacua tion ' is ,assumcd. We used

o~l1y Model 1, beciJ.usc the other m<;>dc 1, wh i ch neeels more parameter

did not converge in <111 Ci.1Scs. Tlw da i 1 Y ril lion i s compulccl <1S

the integral of the contents triJ.j cclory ovar thc' feeding period ••A population based food consumpt~on cstimate was obtained from

the daily ration and length-based growth i.1nd morti.11ity parameters

I' (Pauly 1986)a.

RESULTS•

The time trajectory of relati.ve slomach fullness of the fish.•

is given in Fig. 2. Initial findings showpd that seaCJrasses were

the dominant diet cornponent (Table 1).

Thc time, trajectory of stomach contents <lnd the' estimatec1

feeding parameters are 'given in Fig. 3 and Table 2; respectively.

Feeding time seems to be inverscly rroportional to daylength.

al\. progr<l!n for 1I3~1 compatible !nicrocolllpul<.~r~; callel}' ",r·1i'\XItJlS" ,'imple'mcnting these and related routines [ar cstimatian 'af duilyration and population food consumption Q/8, is available fromthe last author.

4

~---

Likewise, the estimated daily food ration

similar trend.

4-14 g/day) shows a

.e

•

From the 543 fish selected for itomach weight analysis (Fig.

1), the length-weight relationship

Het vieight = 0.014 * Total- length 2.[317

was derived. Based on this, the population parameters provided by.

DeI Norte and Pauly (1989) and' our estimates of daily ration cif.'9.5 9 for an "average" specimen of 27.9 9 vlet wCight, the

population food consumption (Q) over bipmass (B) was computed.

The average daily Q/B of the foui sampling periods was 0.35

(Table 3), i.e. the S. fuscescens population consumes 35 % of its

weight per day. W{th a biomass estirnate of approxlmately 40 t for

the study' area' (DeI Norte and Pauly 1990), this results in a

total consumption of 14 t/day. According to our observation, an

average 64 % of this is seagrass,' i.e. ~ fuscescens consumes 9

t.day-1 of seagrass in the study area, or 0.38 t.km- 2 .day-1,

respectively •

Based on the estimated average evacuation, rate of about 1.8

g.hr-1 , about 84 % of thc stomach contents is evacuated per hour.

• Integrated over a 24 hours cycle, thls leadsto a total

evacuation of about 4.6 9 of asingle specimen of 27.9 9 (to

detritus), and hence to an estimate of egestion of about ,48 % of

the daily ration. Thc rest is respiration, excretion, and growth •

•

5

DISCUSSION

Seagrass has a low nutritive value (Klumpp and Nichols

1 9 8 3 ). C0 n s 1 cl e r 1 n C) .t hat 1 t 1 s t h e ma j 0 r 'f 00 d 1 t e m 0 f S.

fusc'cscens, th1s may expla1n thc rclat1vely' high cstima!:es

obtained far ingestion and egestion rates and, accord1ngly, daily

ration. The f1sh have toconsume a lot of seagrass r~lative to

t~eir bod~ weight in order to satisfy their nutritional

requi.rements. lIence,. they pu t a cons iderable graz ing press.ure on

.the seagrass. llowever, with the high seugrass leaf production of

0.52 9 .m- 2 .day-1 (ash free .dry weight), corresponding to 5.92

t.km- 2 .day-1 (wet 'weight) (R. Rol"lon b , pers. comm.), the

estimated 'consumption by ~ fuscescens

the newseagras~ production.

\iall10unts to only. 6.4 % of

•\Hth the wide cover of seagrass in the' area, . it cannot be'

assessed whether it is the t1 ava ilabilit y ll and not thc

t1 pa l a tabilit y " which guidesthe 'fish to eat, seagrass. Von·

• . Westernhagen (1973) mentioned tha t tbe abundance of food i tem

determines the choice of food, in tha t the most abundant food.

item. is the most likely to be eaten though not necessarily the

most palatable. Bryan (1975) also emphasized that it 1s pointless

to compare food habit· and food preference because the stomach

content of the vlild f1sh reveal not the feeding preference but·

the particular grazing area irnmediately prior to capture. In one

fish sample where seagrass was absent. in its stomaeh, the' green

algne Dictyotn . sp. wns the most dominant component of ·its

bMSI, University of thePhilippines, Diliman, Quezon City,Philippines

6

~\f., ..

contents. Also,' in an experiment done in rearing tanks, the fish

preferred algae (von Vlesternhagen 1973). In this' respect, our

results are also consistent with those'of Palomares and Pauly

(1989). for. Siganus canaliculatus and ~ spinus, who estimated a

Q/U of 16.9 % and 11.5 % perday, respectively. These results are

lower than our overall average resul ts, a~ though the' Hay,' 1990

(12.7%) sampie is within this estimated range. On the other.hand,

the . above fish investigated by Palomares and Pauly (1989) were. .

e' fed on algae, invertebrates and "boiled ~quash", a diet vlhich is

more nutritious. Invertebrates were also present in our field

sampies , 'though .not in great proportion relative to the other

food items.With the frequency of oceurrpnec observed (Table 2),

these inverte~ratcs can bc eithcr ineidcntally ingested witll thc

seagrass, or t~ey may beingested.seleetively to eompensate for

.the nutrients not available~ri sea9rasses; our prcscnt knowlcdg~'

does not allow adeeision on this matter.

...

• As mentioned earlier, there soems tu De an increasing trend

of the daily food ration towards Novemb~r, 1989 and a dcercasc is

observed thereafter until around May, .1990. This apparent pattern. \

cannot be· satisfactorily explained based sole~y' on temperature

ands e a 9 ras s produ e t ion. 'i'l hat wo u I d see m t 0 be t h e . I i k c 1 y

explanation 1s that whieh relates thc feeding rate and the

reproduetive condition of ~ fuscescens. At the moment, this

proposition is being tested through gonadal analysis of the

•

*

sampies .. Gur stomach eontent analysis showed that s.i

fuseeseens is a diurnal feeder. The variation in stomaeh fullness

(Fig. ·2) as eontrastcd to the stomaeh eontcnt weight analysis

7

;~------n

(Fig.3) 'may be exlaincd by the differencc in ,the two approaches,

(volumetrie vs. weight), i.e. a bigger v6lume may have been

accumula ted, though not necess'arily' wi th a heavier biomass • The

overall cdnsistence of thc results oh the dicl feeding rhythm are

further supported' by several 'in si tu observa tions ' of' the f i sh,

during moonless nights exhibiting a scmistuporous state•

resembling sleep similar to the deseription of Randall (1961) for

Acanthurus triost~gus. This is consistent with field observations•

•

"

of spear fishing 'operation where, the, fish lay motionless

("asleep") and camouflaged nmongst the sandy-with-seagrnss

substr'ate thus becoming easy targets for thc spear fishermcn at

night time.

o u r res u 1 t s i nd i'ca te t hat s. .f~~'EQ~.s.Q!!~. i s i nd e e dan

imporfant seagrass, consumer of the reef flat complex ~f Bolinao~

Further studies will aim to. dctcrminc the nutri tional bnsis 'of

this feeding habit and will refine the abovcmentioncd rcsults,

such as those referring to its reproduetive eondition •

•~cknowledgements

We dceply apprcciate thc enthusiastie motivation by find thcdiscussions with Dr. D. Pauly, ICLARM. Dr. E. Gomcz, MSI,facilitated the financial support through grants of thcPhilippine Council for· Aquatie and Marine Research andDevclopment to the Marine Science I~stitute. Our tha~ks also goto the staff of the Fisheries,Stock Assessment CollaborativeResearch Supp6rt Program for their assistanee in same aspeets ~tthefieltl work. He are nlso grat~ful to the Danish InternationalDevelopment Agency for the travel grant given to Dr.P. M. Ali~o

to present thispaper atthe .ICES, Copenhagen,. Denmark.

8

•

.. ,

REFERENCES

1\vila, E. H. (1984). lIormone-induced spawning and embryonicdevelopment of th'e rabbitfish, Siganus vermiculatus (Pisces:SigBnidae). Phil. J.:Sci. 21: 75-108. .

1\ragones, N. (1987). Taxono~y, Di~tribution anel'RelativeAbundance öf Juvenile Siganids and Aspects of the "Padas"Fisheiy in Bolinao, Pangasinan. MSc. Thesis,University ofthe Philippines in·the Visayas, Iloilo, Philippines. 69 pp.

13aga, R. .C'., Sa caya nan, E • ( 1 98 0 ). '1'he fe a s i b i I i t Y ö f cu 1 tur i n9siganids (~lgQnQ~ ~QQQll~Q1QlQ~) in floating cages.Polytechnic Exchange 1: 24-30. . ,

Ben-Tuvia, 1\., Kissil~ G. Hm., Popper,D. (1973)~ Experiments in'rearing rabbitfish (~lgQQQ~ ~lYQ1QlQ~) in sea water.1\quticulture 1: .359-364.

13 r y ci n , P • G. . ( 1 9 7 5 ). F 00 d. II abi t s ,. Fun c t ional 0 i ge s t i v et-lorphology; andAssimilation Efficicncy of thc Rabbi tfish

'Siganus . spinus (Pis,ces, Siganidac) on Guam. Pace' Sei. 29:269-277.

Bwathondi, P. O. J. (1981). The Culture ofRabbitfish Slganus~ in Tanzania.· Internati(~mal Foundation for Science:1-26. .

De La Pa z, R ~ 1·1. , Aragones, N.. V. (1 9 8 7 ). Not e s on t h e ca r 1 y, , developmen t of . S iganus . cana 1 icu la tus. (Bungo Park) in a

floating cage. Nat. Appl. Sei. Bull. 39: 313-318. 'Dei Norte,1\ •.G. C., PaulYr,D. (199Ö).Virtual population

'estimatcs of monthly recruitment and biomass· of rabbitfish,Si~anus canaliculatus, off Bolinao, Northern Philippines. p •.851-854 In Hirano, lI., lIanyu, I. (eds.) Thc Seeond AsianFi~heries Forum. 991 p. Asian Fisheries Society, Manila,'Philippine~~ . .

Hellawell, J.' M.,Abel, R. (1971). A rapid volum9tric method forthe analysis of food of fishes. J. Fish 13iol. 3: 29-37.

Herre, 'A. W., Hontalban, II. R.' (1928). 'fhe Philippine Siganids., Phil. J. Sei. 35:151-185. 'l~~stman~ u. (1975). Some aspects of the mariculture of different•

siganid spe6ies in the Philippincs. Phil. Scientist 12:5-20.Jarre, 1\., Palomares, M. L., So~iano, M. L., Sambilay, V. C. Jr.,

and Pauly, D. (1989). Some Improved Analytical andComparative Hethods for Estimating the Food Consumption ofFishes. Paper prcsented at.the ICES "Symposium on modelsreievant t'o multispeeies management", Oetober 2-4, .1989, DenHaag, Netherlands. '.

Klumpp, D. W., Nichols, P. D. (1983). Utilization of thc' seagrassPosidonia, australis as food by the rock crabNeetocareinu~integrifons (Latreille) (Crustacea: Decapoda:Portunidae). Marine Biology Letters, 4: 331- 339.

Luchavez, J. 1\., Carumbana,' E. E. (1982); Observations, on thespawning, larval development and larval rearing of Siganusnrgenteus, (Quoy and Gaimard) undcr lubora tory )condi tions.Silliman Journal 29: 24-34.

Hangaoang, I. D. Jr.' (1962). Age and ~rowth of three speeics ofSicJ(l,nus in ,the wild and iri ,c'apt{vity: a ,. Comparativestudy b<lsed on their otholi th I S daily rings. t-1aster' sthesis. University of the Philippines, Quezon City. 88 pp., ' ,..

9

l'1ay, R. C., Popper,·D., McVey, J. P. (1974). Rearing anel L'arvalDevelopment of §..iganus canalicula tu~ (Park) (Pisees:Siganidae). Micronesica 10: 285-298 •.

Palomares, 1·1. L·., Pauly, D. (1989). A multiple regression. modelfor predicting the food consumption of marine f1shpopulations. Austr. J. Mar. Freshwater Res, .40: 259-73.

Pa u 1 y, D." (1 9 8 6 ). A Simple Met h 0 d f 0 rEs tim a tin 9 t he F 0 0 dConsumption of Fish Populations from Growth Data and FoodConversion Experiments. Fish. Bull. 84: 827-839.

Randall, J. E. (1961). A Cont:ribution to the.Biology of theConv i ct Surgeon f.i sh of the Ha wa i i an I s land s, l\Ci1n th uru striostegus sandvicensis. Pac. Sei. 15: 215-272.

Sainsbury,K. J. (1986). Estirnation offood·consumption fromfield observations of fish feeding, eycles. J. 1"ish. IHol.29: 23-26. •

Von \\Testernhage'n, H•. (1973). Food preferences in 'cult'ured·.rabbitfishes (Siganidae). Aquaculture 3: 109-117.

Von Westernhagen , H., Rosen thal, H. (1975). Rearing and spawningsiganids (Pisces : Teleostei) in Cl. c10sed seavla ter sys tern.He1g01. w~ss. Meeresunters 27: 1-18. .

Von \vesternhagen, H., Hosenthai , H. ( 1976). Same ilspects' of thesui·tability cf various I?hi1ippine sig'anid species.(Siganidae) for maricu1ture. l\quacu1ture 9: 297-311.

Wood1and, D. J. ( 1 971) ). Rabbi tf i shes neg lected in. Aus tralia areimportant food fish in tropicalcountries. Aust. Fish. 38:21-23.

..

10

·.

List of Figures

Fig. 1. Length frequeney distribution of s. fuseeseens eolleetedat Bolinao, Philippines.

Fig. 2. Relative stomaeh fullness (by volume) of s. fuseeseens atBolinao, Philippines. Standard deviation bars are shown.

Fig. 3. Estimated time trajeetory of stomaeh eontents of S.fusceseens at Bolinao, Philippines. Datapoints are shown.

List of Tables

Table 1a. Details on the number of fish eXClmined per sClmplingperiod ."

Table' 1b. Details on the stomaeh eontents eomposition of §...=..fuseeseens at Bolinao, Philippines. The relativevolume of the food items is quantified in aseale ofo to 5, where 0 = 0-1 %, 1 = 2-10 %, 2 = 11-25 %, 3 =26-50 %, 4= 51-70 %, and 5 = 71-100 %.

Vigurc 1

. }'----- ---- ---

!

I

•. ~

Table 1a. Details on the number of fish examined.per sampling period.

DATE .

August 23-24, 1989

September 27, 1989

November '16-17, 1989

Februa~y 1-2,1990

May 23-24,' 1990

SAMPLES FORSTOMi\CH CONTENTWEIGHT ANALYSIS

108

10

120

SAMPLES FORDIET COMPOSITION

ANALYSIS

33

3

o

o

o

a10 bihourly sampiesbadded sampies during corpusculnr period*see Fig. 2 for further details

Table 1b. Details on the stomach contents cornposition of S.fuscescens at Bolinao, Philippines. The relative volurneof the,food items is quantified in a scale of 0 to 5,where 0 0-1 %, 1 == 2-10 %, 2 == 11-25 %, 3 == 26-50 %,'4 - ~ 1-70 %, and 5 ==, 71 -lO 0 %. '

Olt:crI)mil

24.8. 24.8. 23.8.6-9 9:30-10 10-12

23.8.12-14

23.8~

14-1623.8.16-18

27.9.1f>-:lD

~l'b 123 1.23 123 123 123 123 123

555 35 3

021.302

Rx:d itars~ 455lh:i.d:nt:i.fiEImrly clig:zta:13 3 0

•rrattErl\lg3e , 0 0 0Fa:anini:fura ' 0 0 0~ 000I:Etrib.s 0 0 0Jellyfuh-J.:iY.o: 0 0 0CnEt::a:a 0 0 0U1id:ntificl 0 0 0t::i.ss..e

453

324

000000000000000o 0' 0000

23 5

332

3 3-0o 1 0000000000000000

305

1 ,3 0

432o 1 0o 1 0300o 1 0100

'0 0 1

455

1 0 0

203000o 0 1000000000000

002000000000000000000

1 0 23002000001 0 0000000

84

36

255'43211

4~0

1.7

1.20.20.20.10.10.00.0

a Stomachs were ernpty for the other sampies (20-22 and 22-24 h(24.8.1989), and 2-4 and 4-6 h (23.8.)). 'Total sampie size was13 specimens for cvery point in time exccpt for the interval0-2 h (11 specirnens). In, each case three specimens vlen~ takenfor food i teros analysis, and the remaining ones' for stomachweight analysis. "

.. . ,.' .

Tablc 2. Summary of cstimlted parameters for the food consumptionof Siqanus fuscescens in Bolinao, Phi lippinc·s.

IPERIOD

Aug-Sep 19B9Nov 1989 Feh 1990 May 1990

Ingestion rate .0.9443 1 .0034 0.6691

Evacuation rate 1.4131 1 .401 2. 3.4951~

i'Amount cvacuated ( %) 55.4 .53. G 28.7

'1'ime feeding begins (hr) 04:70 05:00 06:47

'rime feeding. stops (hr) 17:00 18:90 19:45

Daily food ration (g) 11 .61 1 3.95 8.69I,mAN:

Daily populationfood consumption ( %) 3.8 .05 52.32 35.36

HEAN:

0.2943

1 .3669

55.2

05:38

17:55

. 3.589.46

12. 7134.61

..