I --

ICES C.M. 1991 PAPER'C.M. 1991/D : 10 Ref. Sess. U

c

SO14 E METHOLODOLOGICAL CONSIDE R,,TIC S ON TRAWL SURVEYS CARRIED OUT IN WEST AFRICA

bY

CAVE RIVIERE Ala¡ n ORSTOM, CRODT BP. 2241 DAKAR, SENEGAL

A B S T R A C T

Trawl surveys carried out in West Africa, particularly off Senegalese coasts, have given rise to methological studies. The use of Delta distri bution, more adequate than standard mean and variance calculations, generally increases these values. A priori stratification of samplings offers little interest for total catches, or- most of the time for the -main species. Optimal allocations of the ,number of

~ hauls per stratum vary a great 'deal from one -survey to another. A trawl tow 'duration-of-30

- - _ _ -- minutes, is sufficient - - for ab u-n dan ce

- -

- . _ il--_____ & I < . - -, r 7 Fswd~'~ocurngnla~~~ &Is ,O@i

estimations.

- - O 10020833 I -

.-a-% v.... ur;:

;:c!-rrixz 6Pi.J ti,Ì-;l;$z~

:West -'Aft!ica, ,ftomJ Maur,itania-.z,.to ale:l arge rLfaúnistïc .runi ty',< whose des%rLi b-ed -rrbyiwrLonghurs.t n ~ - ( - l 969) .

Specific diversity is high' and a one-hour trawl tow provides

d v a~ tag e sxio f groundfish s zar,ë alllfethe

++-d~tj-+ p. r3-6 TrJ more- interest ti s,t i cs ' are

concerning sampli haul), concerning

2

optimal al locations) and mean and variance calculations. Sor::+ . . results will be compared with those obtained in the i . i r y Coast. Specific research-trips were also made to s t u d y quantitative and qualitative effects of tow durations (one hour, half an hour).

1 . M E T H O D O L O G I C A L P R E S E N T A T I O N O F S U R V E Y S M A D E

I N S E N E G A L E S E W A T E R S

Prospection surveys for demersal resources on the Senegalese continental shelf have been regularly carried out since 1986, by stratified random sampling. The surface of t h e continental shelf between the 10 and 200 m isobathes w a s divided into 1,150 rectangles with 2' longitude and 2.5' latitude sides (5 nautical square miles). This distance is adequate for a standard tow duration of 30 minutes. The rectangles were then allotted into three different areas ~ j ~ i ?

four bathymetric bands. The different combinations Forrned twelve strata.

The total area was divided into the following zones : - Northern : from the Mauritanian border to Almdiss

- Central : from Dakar to the northern border of Gambici ; - Southern : from the southern frontier of Gambia to C a ~ e

Roxo (southern border of Senegal). The bathymetric. bands- were chosen according to p!-!or-

studies on the distribution -of- species. The bands defined NGÌ-S the 10-30,' 30-60, 60-100, _100-200 metre bands ; the lattsr is very small in the Central and Southern areas where, in addition, trawling is very difficult.

One rectangle out of ten in each stratum was choser! 3 ' ~ random (without remission) for trawling. If one of L!IP rectangles thus chosen --proved ,unsuitable for trawling, %he nearest rectangle with the same depth (along the ship's roiite), in the same stratum, was sampled. If the nearest rectangle also

ccprovedt'unsuitable, {another ;~ectangle,-was, picked at E random i r ? the strá'tum. STheir total sLnumberL of: basicc*~~rectangles, -i,;c,,~-qch

I i Stfaturn :andirthez_number 'ofd tcaw.A tows? planned; appe,.a-r ,i

ri3 t dw -st rat i fti 3 KandomqFsampJ i nglr, desi gn; ;;i sJ [ e,t,oi \ I ,rLe+yee est i mat i on var i.ancei .rf i n LJ ,comp a ni son_. 1' w it h ,n,o.nx s,i rat. j,tt i, ed j yap9 om

? 2s ampllli n g:,v 'w h e nt r's t r atat, haveû be en. ap p-ro p.t-:i-a t e 3 y chosen L( _C o 6 h r an y

sdl: 9 7+7f E; G rossJ e i n and : Lau r,ec ,3 $9 82 I:.:{ 3 2 J 81S s3Thef4wi-ghtedc'mean iworked-;öut Caccordj ng~,-~toa.$h~~lnum~e~m of

Point (Dakar) ;

Z.Gk" i ' r07q Ya3 f '73 liuclrf-snrs fi bntj !;g;Î; ...?j i 2

. zv 2 -,/-i u a i': $G - I 9- L r w J - ' 2 L

s r d s f e-l,-,u L x Li F::'s-sGnct.. -d noiJan707ni 3&I:k3t??ijz

t.3-j f # V G 7 .f- ed3 i" e$=?o:. r;r,

, 4 5 3 n3cbj-ia-Y b s f 7 i jS- iSZ) t E i ns3 i -i iJrji4 ) 39 i 'i 3 ;rivo3

atum{ i i and fi3dsr th,ezht_ovtal EbOÎlJ3"Z .f.i i f arr.:32 gr! i 17-753¡;,33

pi- ~-(<xsT +ciTi s , - ~ i ; ~ ~ ~ n 3 3 , E r1-tF-I

5

3

s(Xi) is the mean standard error for stratum i, s(Xi) = si/Gi

Seven surveys were carried out from 1986 to 1990, four during the cool season and three during the hot season.

2 . C O M P A R I S O N B E T W E E N M E A N S A N D ' S T A N D A R D D E V I A T I O N S C O M P U T E D

F R O M N O R M A L D I S T R I B U T I O N A N D D E L T A D I S T R I B U T I O N

Relative abundance indexes and their variances issued from trawl--surveys were often estimated in the past from common mean and variance calculations. This method will be referred to as "normal" in comparison with those in use for particular distribútions.. For a given species, . hauls generally show an . irregul.ar --distribution with many ze,ro values and some very large + catches, and nowadays, the, Delta distribution. system seems best suited to' minimize bias' in mean and .variance caldulacions for trawl surveys (Pennington; 1983 and 1986, according: to the work of Aitchison and Brown, 1957). Delta distribution ',consists in treating positive values' separately with:.'a 's,imple..log-normal distribution; ithen including the' zero va1ueS'i"';A hyper-geometric function, which can easily be computed F L is 'ùsed. The efficiency of' Del ta di stri buti on - depends on' the"number of' trawl 'tows, .on the proportion.,of zero values and-on.'thevariab:ility. range for positive values. (Smith, 1988).

Means (kg/0.5h) and their coefficients of variation ( C V ) , 'compute3 t?$ .Delta!- distei butio-n" ':for the: 'grand' total:. and'rfor the 20 ;;maj n -spec:i'e-s- .-, of groups :: of- .r spec.ies: -'ani Senegalese cónti'nek'tâlsshelf ~betweeñ-~:lO':and~lOÖ meCres. appear.',i ne:, .Tab1 e 2.

i I , , . . _. ,_ - . . - . , . " . <,. . ;;:. . . - . -. . - : A . ' . I - .I i

.-. ._ 2 . ; .i - .

i r I C I

These values concern the^bmeans2'.ofriseven-traw1 ,suweyS.rThe ,100- - 200- ,metre strat-um: issfiotc'concecned- because- i t2'wasf. notisampled

4 trawliiñ'g diff iculEi'es".~~ThS~20.\: spëciessrepresent"fi752!".tÖï-87% Lof : "-th-~~'"$rän~r"t"dt"a'lsi wi'th?FncYrrña1 Tcà1 du:l,at'i'onJ ;3r but'jrth i S rf percentage f= 2 %. : TS canJFy.ilse '&o ~ l 2 9 % ~ ~ ( Sü-tWeji" LS8905:)ucwh'i3hS, Del Ca? d.i stri bût -¡óni_" is used. This highlights an important problem 1 in-kedcwi~thnthe;use

distribution : the sum of the species catches per tow equal to the all-species catches per W w . Delta ion means (Table 2) are generally above normal means, XfÕr StRe kii"ge"y8 g"rÛnt7" BracFydeute"rü3 TaüritOs,3 wh.i c-h is

ö&t Xc&mmo&i Zpzci'eS. BTb2 mea"ns"3- ¿?a& älSoABe bë~lbw Tnormal . nts of variation are a.lso generally higher than The range of the differences depends on the proportion kilue'32:J' buiXn6i2-onl:y3 askmention'ed abÒ3e::iThekseucybathic

has I :few:- j ze r o r á2 g r'ê atG."d i f?e-r e D e2* on7 c$ h el c w h o 1 e3 c'w h e n: r me an s! 2 a re

I & ' b r) b \

' A i.n-':eac~?fi~ûijjvcey::ci nr'stKé S'b-i;thernE-&nd '.(Centrali areás;izdue kt0 i i %*fi P P - C

* .L -

7 7 ¿h?rbe 7:7õtt'i-'iS" ( Red tC p an'd Ô r a )?,Y w h i c h

ZB ,d.rei.ri 3f y3iai'j.Fzjsqai-gt~fcr 47tiriw ~ C E T E 31 , ( s ~ % r

4

computed with normal or with delta distributions. On t h e contrary, the coastal species Pteroscion pe7i and t1:e ~rdep.l:h dentex Dentex congoensis, D. macrophta 7mus show many zsro values and relatively smal 1 differences. The al 1-spec;iss catches, which do not exhibit zero values, have a 4 - 3 % positive difference for the mean, and 18.7% for the CV bei;wigeÏ" Delta and normal distributions, for the seven surveys csmbi~ad, Appendix 1 shows the same results as those presented in 'Table 2, but for each survey, for the species which, on the whole, has the highest positive differences (Erachydeuterus aur i tus ) i the one which has the highest negative differences ( se la - . ; i... $at)

total ) , and an intermediate species (Pomadasis j u b e l i n i P P.. p e r o t e t i + P. r o g e r i ) . For each survey, the differences a r e always positive for 5. a u r i t u s , with means and CV that c m double or more ; this results in high combinate diffc;r.erir;es; The difference values can be positive or negative deperi:lin:-j on the surveys, for the other two species.

Pennington (1983) had al ready noted that n o m " distribution underestimated the true mean variab.i 1 i t , y and therefore gave an over-optimistic impression of the accurac.y of' a. given survey.

. From data on eleven trawl surveys carried out off i;[%@ shores of Ivory Coast, Bernard (1990) made the saro8 calculations for the six main species of the continental s h e l f , In. comparison with the use of normal distribution, B ~ l l ~ % % distribution would reduce the CV by half on average for the a* ix

,, . species, whereas for the five species common to both studjes , , i ( Ga 7eoides, Pomadasys, Brachydeuterus, Pseudoto 1 i thus, Page 1 Tus

-f Dentex), the CV increases in front of Senegal. Bernard naked however, with reference: to Smith (1988) , . that &'!ta. distribution in this conventional use underestimates the !tiem

5

African continental shelf, it is difficult to select a priori strata that present a satisfactory design for the majority of species. The stratifications generally used are in connexion with the shape of the continental shelf in areas perpendicular to the coast (North-South, East-West) and in bathymetric bands assumed to contain the species belonging to the different communities : coastal species, intermediate species, deep continental shelf species.

In Senegalese waters, nine strata were defined between 10 and i00 metres. What are the modifications of the mean variances when calculations are carried out for several levels of stratification, or without stratification? In the latter case, we should bear in mind the fact that, strictly speaking, it is not really non-stratified random sampling because a given number of hauls per stratum is assigned at the start of the survey, thus making the sampling a little more regular.

For the all-species catches,‘ the CV computed by Delta distribution exhibit differences in their variations according to the survey and according to the stratification level (Table 3). Thus, for nine strata, stratification reduces the CV from 15.3% to 9.8% in the LS8709 survey, but paradoxically, it increases from 16.1% to 20% in the LS8614 survey. On average, for the seven surveys, stratification in nine strata reduces- the average CV by 0.9% ; stratification with three depths does not lead to a precision gain ; stratification in three zones: reduces CV by 1.3%. The results are fairly similar if

I computations are carried out by normal distribution (Table 4).

all-species catches. This means that variability- within a stratum is as great as or greater than variability between

:-strata. According to the depth, but also according to the area, r-cer-tain species compensate for other species.

- * r,For catches by species, Table’5 exhibits the CV obtained <:? rom De7ta distribution for the 14 main species, according to rl-n,inec:i strata- and without stratification. Stratification sometimes uces interesting CV- reductions, but for 1 1

(jt specites, out 14, there is‘a negative effect of-stratification > y f o ~ one-: su cat least. .For the three -remaining specjes, CV

qedustkops,go,from,23% to 34% of non-stratified.CV They concern $7 a ~ ,c-xastal u - -i L’., -specie gr:spe,cde_s, theLlred‘. e2LTjagQurGs tryhu

. ~ Bggazstal ,,cdeepc,or efipo.-j-xyJer can :I

l Therefore, stratification seems to be of little interest for ‘

afggt&g 12 gh$-,~,qSsj,f E?-Id3iJ% 79% BEM oc”n” wci Q+ --t Ge?“ - u e - S7milar r

of stratification), were found in Ivory Coast (Cqveeiviere, 1982, 1989) for all-species catches and for those of the two main species, the bigeye grunt Brachydeuterus adritus and

stes carol inensis (Table 6 1. .

A post-stratification applied to one survey or to several surveys for a given species or for a group of species will tend to reduce variance. The difficulty is that it has to be based

. ‘1’

6

on auxiliary information, for example the distribution of water temperatures on the bottom, and not on first-hand data, (distribution of the catches). Otherwise, the estimated Variance can be .made arbitrarily small and is meaningless (ICES, 1990). No post-stratification was applied to our data.

4 . . O P T I M A L A L L O C A T I O N S

One basic rectangle out of ten was arbitrarily chosen to be sampled for each of the selected strata. Consequently, the sampling effort was allocated only on the basis of the surface covered by the stratum. This process is recommended (Grosslein and Laurec, 1982) when previous information about the inter- strata variances is not available. After a survey, it is possible to calculate the way in which the total number of hauls could have been distributed between the strata in order to reduce the final variance in the total area. In general, optimum allocation is obtained by allocating to each stratum a sample proportional to the product of its surface (Ai) by the intra-stratum standard error (si) :

N . s i .Ai ni =

1.sï .Ai

. .For ,the seven surveys carried out off the Senegalese coasts, optimal allocations per stratum were computed in this way' from all-species catches, and according to whether normal 'òr 'Delta distribution had been'used (Tables 7 and 8). I t is to be.noted that results vary a great deal from one survey to and'ther, .,even when surveys are carried out in . the same

'..hydrological. season. Thus, from Delta. distribution and for' 99 . ' ,total, 'hauls, the' optimum al locations, :computed for 'the Northern 60-100.,metre stratum -'vary from ;,4.- hauls, (LS8912) to .45.:'hauls

.-:*.(LS8-6<14) ; -they vary'from' 3, to 40 ,hauls for'the'Southern:'10-30 I * 'met-reL I I- r r , v ,..- %.stratum, - 2 ,.:from ' . 7. to. .36 I for ' the . Central-."lO-30' metre ,-, . ;str,atum.. . ... L"..-. . ._ - T,he: - . , re'sults';'obta,ined with;., the,i.two;'distri . . bQ"ti'0n.G 'for a .: -I.- . ,-g.i,ve,n.?survey \I" 1 1 , ' - ~ . . I

can ,diffe . ... 1 . '-great :,:deal:- 1 - LS861%'"and, ?-, - - .LS891'2 1 . - Et ;:: !;w o;uJ d..$ :~. :seem ~ ,w i se. , on ... e,': . who! e to.'' - somewhat"'' . , - __ _-. - _ _ -. . -. rekiÚc"ë- !-, the ezal z r t o . b r , 1 ocat j,o ed -';on I -the: 3 0:6 O,:' ,_me e, I y.stwrata* 'and EoVs:.'.i'riYrease

computed and, considering the variabilit'y' Öf'"'válues~~~it( Se?" sE~re~~ra,bJek, to.. ,keep- the.,+,al loc*a$ions al ready in use ~ for future

("%$t/-JF,ge_ [USé :- dftën tfie;. 60-1'0.0: -I ,:k1 ogèJa mette,":s%.r :@ol-tth8;:-:avéK 'id: -Thecal l'ocâtionsfinFÜse, e y rsp t. i.ma 1': F 2 1 q O I C ã t i sons however','- - i' , , ,:: f * ,,*e i-

li-. %%,*"&/5 i c 4 c;-; sz+.*.J; j , ,%2$225," , - , ;sr; . ;TrZ Ø

1 IL.',! 171 ;JjrrlLiay $3Si.r. ( f - ; O ~ ~ < ; . - ~ ~ * 2 - ' .*i 1 1 r j s rSa ,-lo &qgy$ey5 y5.3 3 $pz,,>z , ,*. .- 7 .- 0 ~ 3 943 7s esori,? ?o'(- t.ri.6 ~ ~ 1 ; 1 7 ' ~ v ~ ~ 3 as.ifiz,qE-- r i r. SBE r &'!*i,? %U3 5. YU B Z.t&"r 2.3 Ij" kvrC&z a r 3 3 c'i jJ 7; jz a .. '

7

5 . P R O P O R T I O N A L I T Y O F C A T C H E S B E T W E E N H A L F - H O U R A N D O N E - H O U R

T O W D U R A T I O N S

5. I . THE PROBLEM

Trawl surveys are often carried out with a standard half- hour tow duration. This offers several advantages compared with longer tows :

(!)-a greater number of tows can be made in one day, which will reduce the duration of a survey when the number of tows has been Dre-established (one basic rectangle out of ten, for example) ; survey costs will be lessened ;

(ii)-quantities to be sorted out will be will save time for the men in charge and could fewer men on board and/or more time to spare sampling.

while a trawl is operating. Direct observations One possible drawback could stem from

reduced ; this lead to having for biological

fish behaviour have repeated1 y

shown that many species swim in the mouth of the trawl for some time, until fatigue makes them swim into it and be caught.' This 'swimming time could depend on the species and on individual length. Wardle (1986) noticed that larger fish, such as adult saithe, cod and haddock, swim for very long periods in the mouth of the trawl. Let us suppose that-adults belonging to an evenly distributed species can swim for 15 minutes in the mouth

',+i ...o f.* the trawl : a half-hour tow duration will only catch half " - the',individuals present on the passage of the trawl, whereas .a one-hour trawl tow will catch three out of four. There will .be no direct proportionality between tow durations and catches. On the other hand, if direct proportionality does appear, it will be possible to assume that a half-hour tow..is , ,. just as good to sample a population as an hour's tow or'more.

sThis ris ... what- we.-intend, to .study for. ,species. belonging to -:demersal communit,i,es. of West Africa from 'Mauritania'. to "Angola, -.based on ; pai rs of h,al f;-hour and: ... . . , .one'-hour:, . -~

. 1 . trawl',, . I < . . . ~ . tows c

(3 2 E 2. [> c ;y,,:, .; c-! i nt,fr;ontiLof: :-S,enegal-.. . . . . I. -. . $,' c4 7 - . " ._ ._ . .

I . . ,c...,.-

. ,

a

E i g h t

28-31 14-17 25-29 01-04 01-04 13-15 10-13 10-13

research t r i p s were c a r r i e d o u t :

January 1987 December 1987 January 1988 June 1988 June 1988 February 1989 A p r i l 1989 Apr i 1 ' 1989

Louis Sauger

,

NDIAGO Louis Sauger

NDIAGO

8 p a i r s 9 "

1 3 (I

7 "

7 "

7 "

7 "

7 "

65 p a i r s

A l l t h e tows were c a r r i e d ou t i n broad d a y l i g h t . The t r a w l on board R.V. Louis Sauger i s a high-opening t r a w l w i t h a 27 metre headl ine and a 45 mm mesh opening s i z e i n the cod-end. R / V Ndiago has an I r i s h model w i t h a 45 m headl ine and 6 0 mm mesh s i z e i n the cod-end. Trawl speed was about 3 . 7 knots f o r both research vessels.

The d i s t r i b u t i o n of t r a w l p a i r s accord ing t o depth i s shown below. I t runs from 18-19 metres ( 2 p a i r s ) t o 120 metres (5 p a i r s ) :

s t ra tum depth 18-39 m 40-69 m 70-120 m

number o f p a i r s 26 22 1 7

Catches per p a i r f o r t h e a l l - spec ies catches and f o r the main species are shown i n Appendix II. Double absences, showing . t h a t a p a i r d i d n o t take p lace i n the d i s t r i b u t i o n area o f the species, a re n o t mentioned.

- I .

I _ 5 . 3 . PROCESSING AND RESULTS ~ ' /.

e - - 1 . .. b - < , - - t

- . The-- use: -o+: rat?*os- :seems- )su i ted 1 t o ' the,'^data L (Cochran, n t f e r , , 1983,) h& - r a t i o estimate" * R = Y / k (one-hour / " ha l f -hour chF"'hea'h) 'has-h'ay v-a-ri ance lwhi ch can be

approximated if n > 30 (below t h i s , t he apprÖximafi-on'usedr7is unsu i tab le and b ias becomes t o o g r e a t ) .

The r e s u l t s f o r t h e a species catches and the main species a re g iven i n Table 9. Only 3 "species" o u t o f 23 do no t

9

include value 2 in the 95% confidence limits of the ratio and. these values are exposed to bias because n < 30. The ratio estimate for the all-species catches (1.95) is very close to 2 ; a similar ratio estimate (1.96) was computed by Barnes and Bagenal (1951) in the North-West Atlantic for the same durations. In the North-East Atlantic, Pennington and Grosslein (1978) showed that fifteen-minute tows caught proportionately as many eel pout and haddock as two-hour tows. Separate results for R / V L.Sauger and R / V Ndiago are given in Appendix III. It may be noted that the R va.lues for R / V L.Sauger are close to those of both vessels combined (Table 9), and that the values for R / V Ndiago (when catches and/or the number of pairs are not really too small) follow the same direction ( < or > 2) as those of R . V . L.Sauger ; this justifies, a posteriori, the grouping of results from very different trawls (high-opening and Irish models).

In order to have over 30 pairs in each stratum, the all- species data was arbritarily divided into two depth strata : 18-44 m and 45-120 m. The results are given below.

18-44. m 45-120 m

n 33 32 5 *T

Y 804.5 k X 449.9 k R 1.79 2.14 95% Confidence limits 1.44<fi<2.14 1.52<@<2.76

036.4 k 390.7 k

- -

( . - 7 , , . ~.

. .The "ratio estimate fi is sma1.ler. in ,the" coastal .fishing .grounds :_ than: : in the deeper . _ , grounds.. . I Considering that .' the confidence ~ -.intervals .. ove,rlap, the,' two ~. values. ! 1 . are.' not significantly-. :different.-. Value.. " . ..... .'2:. ~ - . . . is. included ,,.in.-. these i nteryàl s. ,:,: . .

is:.;_rather:- diff.icu1.t .tÖ':; imagine- justif,<cations::'fÒr. a ratio:+estimatg.-; of less., than '2i -be'tween' one-hou'r.. and -halfihour trawl tows, except in" the' . casé _ _ _ i ro'fb,. triwl.. . i:- I ... c 1. satu'ratiori,"Á ,:<est 'was carried out to see if such a satiiration cdu1d'be'"foÜnd"with the dat,a. In order to do this, the ,% : ..,L. :-.data I -,was I d,ivi-de,d i,nto two series : one for which the'sum of catcKes pef pair was'inferior to , I O00 k (lowest walue = 113 k), and the.~,ot~~r$_,o,T.~.wh~i~h L ?e this sum was above 1 O00 k (highest . va.lue.,=. . .t:LJL*:, r 1 - . . t ,5.,,.7$8 :la4 k')<",'The- i i d d s.:: = results 7. are g i ven be 1 ow. f

. . .. . . . . - . .. . , , . . - . ( . .

,.l.l ' ;..It

zeubizs-3 9d3 3arj;l n

f rpp i- &p.",* r^i ividuals with@-a

portionaly as often caught by the case :

10

(i)-Among the larger species, the white grouper fpinephe7us aeneus is the only one regularly present (n=47). The ratio estimate is 2.0 and the larger individuals of this demersal species are equally well sampled with one-hour or half-hour tows. Thus, in 1987, seven E.aeneus with a fork length of over 70 cm were caught during the half-hour tows, as opposed to nine during the hour-tows.

( i i )-Large barracuda (Sphyraena a f r a ) , over 1.8 metres in length, probably belonging to a school were caught-during a half-hour tow. We should bear in mind, however, .-that other pelagic species, with a high swimming capacity, have a ratio estimate R largely above 2.0 (particularly the horse mackerel Trachurus spp), although this value is included in the 95% confidence interval.

Godo and a1.(1990) observe that 5-minute tows are at least as effective as longer tows (up to 2 hours) to catch fish of all sizes in North Atlantic waters, even when, owing to small fish/large fish differences in swimming capacity, a relative decrease in catch rates of large fish was expected with decreasing tow duration. They suggest an interesting idea to explain this discrepancy from expectation. The trawl may have higher efficiency due to a surprise factor during the first few minutes of a tow, before a school is established, inducing an alert reaction at an earlier stage in the catching process.

Lastly, the linear regressions of y on x were camputed foi the 23 main species or groups of species after a

I logarithmic transformation of the variates in order to stabilize variances. The greater the non-transformed values, the greater the variances. On account of the use of logarithmic transformation, and because calculations are difficult, linear regressions were not used to study direct proportionality between one-hour- and half-hour catches, whereas this question was -pr_eviously studied with ratio estimates, simpler to use.

- - as used : - 1 .

. _ t 2 - * 1-1 : I

- , , . . , . . .. . ". . . ,

11

,

-1 '

I

8 .

5.4. CONCLUSION ON TOW DURATION .

The use of half-hour trawl tows appears to be sufficient for abundance estimation in a given place. Results can be doubled to estimate catches per hour.

C O N C L U S I O N

The use of Delta distribution to compute abundance estimates and their variances from trawl survey data generally increases means and their coefficients of . variation. Nevertheless, the use of this, distribution can be recommended because the results are unbiased, unlike normal calculations, provided the number of samples per stratum is not too small. There is , .however, a disadvantage : the sum of abundance estimates per species encountered is higher than the all- species abundance estimates.computed from the total catches per haul.

When one-tenth ' of .,the. basic rectangles is samp'led ( 5 nautical square. miles), the coefficient -of variation of the all-species abundance esti.mate,on the continental shelf in West Africa is approximately 15%, which is a very reasonable value, ': according to studies on the subject.

Stratification of samples according to .areas and, depths, supposed to reduce variances of abundance estimates, is of .

1,ittle- interest for all-species, catches, in Senegalese, or Ivory Coast waters and, doubtless . . . in the whble-of West ,Africa. In the different strata, certain species caught 'compensate for others. Stratification . may ~ be; interesting' for some . . particular . , , species,

..... . . ,

. _ , ..'

... _ . ~. . -:.- * . . - but-,it is-not :a, general" rule. i.:. .. r _ _ T . ; ,

.- ,.%. . _ - * , "

'; ; i c, , : : A posteriori computations of

s t rat u m 3 o r i. a T1,- s p e ci-e-s;: i c a t .ch e.s ;!var survey;;.to.,:another ,. an,d,!~i.tiz~ap~ears~~:~,o sat.i..s_fa.cto.~y;,.~.,t~~.~~~~o~~~~~e -_,_ 52-E~ 1 .yy stratum surface only.

12

R E F E R E N C E S

AITCHISON (J.) and BROWN (J.A.C.), 1957.- The lognormal d i s t r i but i on . C-i"..r-idg.e. Uni.Y-., .._.. P-ra.s.s-, .."...._ La.n.da.n , 1 7 6 P .

BARVES (H.) and BAGENAL (T.B.), 1951.- A statistical study of variabi 1 ity in catch obtained by short repeated .trawls taken over inshore ground. J.> _...._ B~,.K ..,..-. B..~?LL ..._. UAL, 29 : 649- 660

BERNARD (Y.), 1990.- Sur l'utilisation des campagnes de chalutage comme outil de gestion des stocks. Cas de la pêcherie démersale de Côte d' Ivoi re. ~ ~ . ~ ~ L . ~ - ~ , ~ ~ , . ~ ~ . t , , u ~ ~ . Q.c.~~.~.Qs..r-~~~h-i.~~~.~, sept. 1990 28 P . miméo.

CAVERIVIERE (A.), 1982.- Les especes demersales du plateau continental ivoirien. Biologie et exploitation. These Doct . Etat Sci . Nat. , ~ - ~ ~ ~ - - ~ - ~ ~ - ~ - ~ ~ ~ . - ~ ~ . . - . . ~ ~ F a c ~ - ~ . ~ . ~ - _ d . e . ~ ~ . ~ . . ~ ~ ~ . ~ - ~ , ~ - - ~ - ~ . - . " . ~ - ~ ~ . ~ . . n ~ Y 2 vo 1 - -7 4 15 et 1 60 P

i

13

PENNINGTON (M.), 1986.- Some statistical techniques for' estimating abundance indices from trawl surveys. .F.-~-.s~...,,,, .@-U,..l,J..e,..y vol. 84, 3 : 519-525.

PENNINGTON (M.R.) and GROSSLEIN (M.D.), 1978.- Accuracy of abundance indices based on stratified random trawl su rvey s . ~ - ~ - ~ ~ - - - ~ n - ~ ~ E x ~ - L 2 ~ _ . _ ~ ~ . I _._. 3&..&-M.=ee CM 1978/D : 13, 34 p.

SMITH (S.J.), 1988.- Evaluating the Efficiency of the delta- I

distribution mean estimator. .&kxng.$.c¡-~~., 44 : 485-493.

WARDLE ( C . S . ) , 1986.- Underwater observation technologies. Present status and review of results i n fish a n shrimp behav i our research . ~ ~ ~ ~ ~ ~ ~ - . - ~ - ~ . ~ ~ - ~ ~ . ~ ~ . ~ ~ - - ~ ~ - ~ ~ . . ~ . . ~ . ~ . . ~ . . ~ - v . ~ . s. b . - ~ . ~ , . m ~ . _ . . . . . i ; _ . ~ ~ . ~ - ~ ~ - ~ ~ Q ~ ~ ~ ~ . ~ ~ ~ . n , . . . . . . - M . a x . i c s , 24- 28 nov . 1986. F I 1 : ExCons-STD/86/D,P.lY 55 p.

' . ., . _ .

i

14

f I \ DEPTH 10-30 m 30-60 m 60-100 m 100-200 m ALL

AREA DEPTHS

Northern 76 (8) 71 (7) 103 (10) 97 (10) 347 (35)

Central 137 (14) 127 (13) 98 (10) 42 (4) 404 (41)

Southern 214 (21) 131 (13) 32 (3) 22 (3) 399 (40)

-~ _ _ _ ~ ~ ~

A l l areas 427 (43) 329 (33) 233 (23) 161 (17) 1150 (116) i

f I I 1 I I I I \ DEPTH 10-30 m 30-60 m 60-100 m 100-200 m ALL

AREA DEPTHS

Northern 76 (8) 71 (7) 103 (10) 97 (10) 347 (35)

Central 137 (14) 127 (13) 98 (10) 42 (4) 404 (41)

Southern 214 (21) 131 (13) 32 (3) 22 (3) 399 (40)

A l l areas 427 (43) 329 (33) 233 (23) 161 (17) 1150 (116) i

TABLE 1 : Number of basic rectangles per area and bathymetric band, and number of planned trawl tows (in brackets).

. 8 ..

P

15

Brachydeuterus auritus

Trachurus + decapterus

Dentex angolensis

Pagellus bellottii

Boops boops

Chloroscombrus chrysurus

Selacian total

Cyrnbium spp.

Galeoides decadactylus

- .

" , _ . I

Dactylopterus vol ïtans

Dentex congo. + macro.

Pomadasys spp.(-inciaus)

Pteroscion peli - .- - -. . . -_ .

---I- I . __ ., .-. Arjus.spp".".;: ~ ' ! : . , .

.- ---_- - . . - -_ - __ , Sparus caeruledstictui

Plectorhynchus medit-.-:.--- I'

Scokberi jap6nicus.Z: '~ E . . -7 Pseudoto1ithus:spp.

Epi nephèlus+Mycteroperca-

. . - . ; i 7 ":

I I

I

.: . : i - ' . -* ' --

1

I '

&,O 1 e.er 1 S.St 5;. .. t ' ' - ' ' - . i . . .

GRAND TOTAL CATCHES

- x No

- 58.8

56.7

76.1

39.7

74.2

78.6

14.7

71.7

76.9

64.9

94.3

71.4

87.1 - - 76.2 -- 63.4

7d. 7-

81.3

8 9 . 5

1 I

-i-- 1 t.

I 55; 8-

76.5. !

0.0

-

-

DELTA MEAN

94.8

81.1

73.5

32.3

31.8

25.7

19.9

16.4

12.4 - 11.3

8.4

7.7

-: 7.0, -- -

-* 6.8, - ---

6.3

i

-5.0

! 4.9

i ,4.9

! T ' t --$4.9-

i - i

i 14132 __

450.9

MEAN DIFF.

40.3

14.2

21.2

7 . 0

12.1

8.0

-3.2

o. 9

2.8

-0.6

-0.2

-0.3

O. 7

-' 0.8 . .

- 1.7

- * 0.1

'-1 ..l

'r 0.8 1 -- -._

1 s.; 1 !:? -- 21.0

X MEAN DIFF.

+73 J9

+21,2

+40 5

+31 ,8

+61,4

+45,2

-13,9

+5,8

+29,2

- 5 JO

-2,3

-3,E

- .

+11,1

- ELTA cv - 46.1

33.9

56.0

28.4

54.5

61.1

22.4

35.2

51 ..7

42.8

72.4

37.8

42.9

38.1

- l _ .

39.4

35.8.

51.0

36.1

26.8,

44.5$

__I

14.6 -

cv DIFF.

18.3

8.2

18.7

6.7

7.2

11.4

-4.9

5.3

9.7

-0.8

4.4

O. 7

2.5

4.6

x cv DIFF.

+65,8

+31,9

+50,1

+30,9

+15,2

+22,9

-18,O

+17,7

+23,1

-1 ,8

+6,5

+1,9

+6,2

+13,7

--I - I +38,3

+ 4 *L

-1,4 !

- 4

. . - i - . -1

i

1 -- +30,8 i

.+2 1 ,6

+18.7

&Í.$F& "'"i -II

:ß: 3 ~ m r +ar i at i on

percentages.

16

I

STRATA LS8614

9

20.0

Í- 3 areas , 3 depths

3 3 depths 23.9

GRAND TOTAL CV'(10-100 M) " cv

e 3 areas

15.3 14.7 15.1 14.9 15.5 16.8

TABLE 3 : Coefficients of variation of the grand total mean (10-100 m) per survey for three stratification levels, and precision gains (CV reductions) compared with non-strati f ied values. Del ta distribution.

LS8905

64.0

53.6 ....._... ..*

37.5

47.2

40.2

41.4

LS8912 LS9002

33.6 67.7

52.1 54.1 ......... "....

34.7 25.2

54.1 41.8

72,:,_6. 31.5 65.1 36.8

LS8717

32.4

59.9

38.1

55.5

89.8

65.5 I I

I

22.1

39.1

81.7

66.3 --

48.1

53.6

LS8806

45.8

53.4

23.4

42.1

49.9

54.1

22.5

29.5

52.3

58.8

69.9

76.6

Trachurus i Decapterus

Dentex angolensi s

I 9 51.5 27.1

1 70.5 '46.5

9 62.2 38.5

I 77.6 45.2

Pagellus bellottii 9 39.9

1 47.3

21.6

39.2

60.9

55.0 _..___

78.6

74.2 ...-."-

40.1

44.0 __-

42.6

57.9

65.5 38.9

56.7 49.7 -.._... -. ....

48.6 ,553- , 63.0 53.3

44.9 26.7

43.3 29.9 -I-.---

34.4 9.(.5 , 45.9. 76.1

Boops boops

Chloroscombrus, chrysurus

Cymbium spp.

Galeoides decadactylus ~ - I *-

. -

9 40.9 41.1

1 91.1 49.3

9 74.4 53.1 1 66.5 53.7

9 28.6 36.4

1 39.9 37.1

9 41.3 2%: 1 - - 45.6 47.8

48.4

53.0

54.1 -..-I

21.3

21.7

39.8

17

cv lEOUCTIOh

~

Brachydeuterus auritus 636 12,5 X

17,2

33,7 x

-0,9

-1,6 X I I

38.0 36.7 18.0

38.1 1 38.1 i 26.0 834 22,7 %

I 1 I I

7,1 11,3 X

~

312 8,4 X

Op3 0,6 X

42.0.

34.8 -- ......... 41.4

1, 5 1,. 4

36.1

40.8

-3,l

-7,7 x

oJ9 1,3 X

100.0' 38.2

44.8 68.5 1 - 90.5 9jr 7

_ _ I-- 34.9

37.1

2 59.4

61.0

. , i 3 I - ' I 8 - , s 74.9: 151.4 ' I 11 :

1 I . . . . . . . .

31.1 36.2

30.8' 41.9 -

I

45.2; 44.f

54.2, I ( 51.5

t 30.5i 31.5

t I 33.3; 37.0

1 l

25.9 1

36.31 31.7

I 1

2i.4 I

,

2 1 7

6,8 X

13,7

24,2 X

3,5 8,4 x

/

56.2

58.6

X l d k 32.6

34.9

3 main iverage Delta

18

~

Para- Brachydeuterus meters auri tus

MEAN 40,3

CV 32,2 X

MEAN 40,4

Balistes ALL-

carolinensis SPECIES

CATCHES

15,l 194,O

26,2 X 11,8 X

15,O 193 ,O

, 6

(3 areas and 2 depths)

3

(3 areas)

2

(2 depths)

1

(without stratification)

cv 32,l X 26,5 X 12,2 X

TABLE 6 : Means (k / 0 . 5 h ) and their coefficients of variation for the two main species and for the all-species catches with several stratification levels. Ivory

(. L -Coast continental shelf (Survey CHALCI 79.01).

MEAN , 4 1 ,O

. , I

15,2 195,o

8

MEAN

CV

h

40,6 14,9 193,6

32,s X 26,2 X 12,8 X

a

CENTRAL

19

SOUTHERN I AREA 1

60- 10-

lOOm 30m

10 21

10 13

10 20

9 20

8 32

10 12

8 28

27 5

11.7 18.6

STRATA

ALLOCATION USE0

30-

60m

13

36

3

4

2

5

15

4

9.9

LS8614

LS8709

LS8717

LS8806

LS8905

LS8912

LS9002

10-

30m

- 10-

30m - 8 .

30-

60m

3

6

3

10

11

3

5 -

307

60m

7

2

8

6

3

NORTHERN

60- l O O m

10

17

21

4

14

11

9

49

18

25

16

11

7

21

I

3

10

8

5

6

5

2

9

30

LS MEAN 5.9 5.0

60- I OOm -

3

15.3

O

1

1

1

8

I

8 - 4.1 -

* ,

ïc-:' ,-

TABLE 8 :- Optimal alldcation of trawl tows per stratum for each

20

Û

1.95

i

Sd

0.18

SPECIES

2.39

2.10

0.92

2.30

1.26

3.32

1.66

Grand total

Trachurus spp.

Decapterus rhonchus

Scomber japonicus

Boops bwps

Brachydeuterus auritus

Sphyraena spp.

Dactylopterus volitans

Pagellus bellottii

Sparus caeruleostictus

0.46

(1.34)

(0.92)

0.62

(0.18)

(0.80)

(0.25)

Dentex canari ensi s

Dentex ango. + macro.

Epihephelus aeneus

1.82

1.73

1.66

1.44

2.03

1.69

K 8 9

3.53

2.44

3.93

1.18

2.14

1.62

Total groupers

Pseudupeneus prayensis

i ~

0.26

0.19

0.27

(0.46)

0.53

0.38 , -

0.31-

(0.22)

0.45

(2.69)

(0.25)

0.43

0.20

Driacanthus arenatus

’lectorhynchus medit.

Jmbri na canari ensi s

Jseudotol i thus spp.

Zeus faber

?aja miraletus

lustelus mustelus

Sepia spp.

- N -

6f

45

1 5

1r

3 E

2E

2í

2€

52

3í

3 8

2 8

47

5c

41

26

3c

27

1 4

46

56

21

55

-

5332:

9001

601

1 1 4

928E

220E

212

2655

4282

754

295

5507

465

596

’ 676

3710

594

167

132

307

431

800

427

-

TX -

!7347

3763

289

125

4042

1755

64

1599

2347

437

180

3830

229

354

357

1051

243

43

112

144

266

285

192

-

820.4

183.7

32.0

8.2

244.4

78.8

9.7

102.1

82.4

23.6

7.7

189.9

9.9

11.9

16.5

142.7

19.8

6.2

9.4

6.7

7.7

38.1

. 7.8

- X

420.7

76.8

15.2

8.9

106.4

62.7

2.9

61.5

45.1

13.7

4.6

132.1

4.9

7.1

8.7

40.4

8.1

1.6

8.0

3.1

4.7

13.6

3.5

? 2 l r 9 6 Sd

1,60

1,49

(0)

(0)

1 ,O8

(0,911

(1,751

(1,171

1 ?31

1,36

1,13

(0,541

0,99

0,95 - - _ I .

1,28

(3,101

1.56

ro i

(0,691

1,30

1,23

[1140)

1.49

and ratio estimate (6) with its standard deviation (Sd) and 95% confidence limits for N pairs of one-hour ( y ) and half-hoúr catches (x) of the main species. ( ) skewed, n < 30.

21

Grand total

Trachurus spp.

Decapterus rhonchus

Scomber japoni cus

Boops boops

Brachydeuterus auritus

Sphyraena spp.

Dactylopterus volitans

Pagellus bellottii

Sparus caeruleostictus

Dentex canariënsis -

Dentex ango. + macro.

Epinephelus aeneus

Total groupers

Pseudupeneus prayensis

--i - _ _ _ _

i Priacanthus arenatus

Plectorhynchus medit. i r

i thnbrina i * canariensis i

f 1

wsteíus mustelus I -I

- N

-

65

49

19

1 4

38

28

22

26

52

32

39

29

47

50

41

26

30

27

c -Y

14;

4 6; i I

561 Y L

%C 212

.G .z 55$

- :el &i c

SCOPE (b)

0,826

0,796

0,183

O, 366

0,905

0,870

1,103

0,903

0,799 i

0,835

O, 575

O , 679

O I 560 ..

0,498,

-, 0,926

. t 0,780

; 0,683 -i - . .

c y

INTERCEPT

(a)

0,710

0,636

O, 930

0,434

0,457

0,139

O, 224

0,311

0,495

0,396

0,364 - -

0,813

0,411

-- 0,437

0,276

0,332

0,598

0,449

0,255

0,334 i ..f f

0 , 2 9 0 ~ J- * f

0,644 c 1 -

CORRELATION

COEFFICIENT

0,87***

O , 70***

0,17

0,36

0,80+**

O ,74***

O, 79***

0,85***

o ,80***

O ,81+**

0,54***

o,a7***

- __ - - .- -

0,49***

O, 43**

0,84***

0,67***

- ,' 0,62***

#i

I ' ~ -:

Figure la : Linear regressions, after log + 1 transformation, of one -hour

-. . . . . . . . . in

,E

t i

I catches on half-hour catches for the main species or groups of species

tu N

. .

I one-hour'catches on half-hour catches for the main-species or groups of species

N P

N UI

i es

28

LS8614

LSE709

LS8717

LSE806

LS8905

LS8912

LS9002

ALL LS

BRACHYDEUTERUS AURITUS

DELTA MEAN X MEAN DELTA CV x cv MEAN DIFF. DIFF. cv DIFF. DIFF.

66.6 16.2 +32.3 X 37.3 12.8 +52.2 X

185.6 97.. 4 4-110.4 X 41.6 17.7 t74.1 x 74.5 17.6 +30.9 X 32.4 9.2 , +39.7 x

127.5 61.2 +92.2 X 45.8 14.7 +47.3 x 134.6 74.9 4-125.3 X 64.0 31.7 t98.1 x 47.4 1.6 +3.4 x 33.6 1.6 +5.0 x 27.2 14.2 +109.2 X 67.7 40.5 +148.9 .Y

94.8 40.4 +73.9 x 46.1 18.3 t65.8 X

L.58614

LS8709

LS8717

LS8806

LS8905

LS8912

LS9002-

r3

JI 1.

v a l ues.

SE LAC1 AN TOTAL

DELTA MEAN X MEAN DELTA CV x cv MEAN DIFF. DIFF. CV DIFF. DIFF.

24.4 - -1.3 -5.1 x 15.8 -4.2 -21.0 x 3 17.1 3.9 +29.1 X 26.8 8.9 +49.7 x

36.2 -24.4 -40.3 x 39.2 -30.5. -43.8 x 15:7 -0.6 13.4 x 21.5 -7.1 -24.8 X

12.3 o. 9 +7.4 x 16.3 4.5 +38.1 x 19.1 O. 7 +3.8 x 21.6 -1.6 -7.0 x 14.2 -1.3 -8.1 x 15.3 -412- I-- -21.5-X

,

ALL is 19.9 -3.2 -13.9 x 22.4 -4.9 I

-1s:o x

30

8.8 4 .5

285 101

104 32.3

O 10.6

O 0.4

O 22.8

0.3 o 4.9 1.8

52.0 5.6

2.6 2.5

j i i

i i

1 i i f

- 8

15 .2

102

453

442

12.3

3.0

1.4

O

6.6

SPECIES

7.2 8 .5

106 442 BRACHYDEUTERUS

13.9

, 460

AURITUS

I

366

260

2.3

O

0.5

0.3

O

0.8 4.1

7.8

O

9.9

9.8

2.0

71.4

13.9

16.4

6.7

O ~

100.3

3.0

1.7

O - 0.8 O

3.5

35.2

SPHY RAENA

SPP.

20.0 45.4

20.0 14.6

2.0 3.1

1.8 0.6

o 1.8 0.6 O

3.5 3.0

6.6 4.6

41.5 21.3

o 1.7

5.1 3.9

o 2.3

139 36.6

5.8 1.7

178 123

66.2 2.9

42.0 35.8

55.2 35.5

501 155

55.3 63.2

i 1.4 12.2

45.4 12.2

~. ~ , ,

,19.3 4.,1

7.2 0.5

16.2 9.7

1.1 1.1

27.7 8.8

i

DACTY LWTERUS

5.1

64.5

27.4

1.5

12.0

5.3

4

VOLITANS

4.4

119

45.6

17.3

1.4

3.8

PAGELLUS

1.4

5'1

1.1

BELLOTTII

0.1

0.4

7.6

0.2

079

645

2.5

10.4

O

113

822

445

17.7

71.8

64.7

10.2

42.4

17.9

- 4.0

25.2

22.5 * i-

''35.5

y.4.0

CANARIENSIS

0.1

o 0.4

o

579

565

1.1

5.9

o. 2

108

701

167

9.7

32.2

.31.3

23.5

29.9

16.7

0.7

18.5

15.4

a ' 2.9

9.7

33.5 o 9.9 o 7.3 2.2

39.7 58.6

243 130

1.1 33.3

39.6 15.3

55.0 30.5

12.2 8.0

O 5.9

10.2 o 44.3 68.9

36.0 44.9

0.7 3.2

2.2 0.9

is

,2.2 7.0

7.4 2.8

3.7 f 0.4

3.1 1.6

1H

2.1

2.5

5.7

7.8

54.9

74

6.3

73.6

37.7

42.1

4.8

3.7

138.8

7.6

6.8

14.6

. 1 . 6 i 10.4

1 10.3

1 /2H

95.1

70

1.7 O

7.0

1.1

1 5 . 7

O 107

o. 5

2.3

268

73.9

1.a O

O. 4

604

56.5

3.5

25.2

78.4

124

42.5

9.2

1.5

12.8

85.4

131

* 46.2

1.3

4'. 3

7: 7

:.o:$

i -?,+- 3! a

siita 0 2 4

f 2i.2

8.6

-2'f3

2.9

1.4 :q3.*5

3.qlJ$2

10.2

6.7

37.6

0.9

3.8

0.6

3.7

0.3

42.5

O O

39

49.7

o 3.2

O

204

32.4

0.3

5.8

56.6

41.9

14.8

O

o. 1

2.1

27.0

90

36.1

O

2.7

3; 8

* $0.9

I .'1.3 * , % y

13.. O

i. 1<8

8.3

O -0-

O O. 7

12 1 7c -

I .*,

29

1575

261

SPECIES I-- 1172 199

AL L-SPEC1 ES

196

315

762

132

259

57

287

207

CATCHES

. TRACH URUS.

160

336

489

56

226

56

162

54

TR-ECAE - -

524

977

739

713

353

!313

475

189

1079

64

0.7

283

83

61.9

187.7

-79:5

55.3

2.3

294

438

297

329

199

2083

212

84

355

9-

0.3

43.5

46.9

O

63.0

84.4

28.1

2.7

1H

387

748

316

1 /2H

141

370

277

B 9 0

!318

92

342

646

351

920

137

754

496

148

205

527

1269

246

60

132

152

178

88

O

2.5

304.5

o. 2

4.1

9.0

O

O

: 55.6

- . o : .c 440

8441

t 69: 1 O

56.1

208r F 128.5

5.6.8

r.2 - CI . b-

(z

O,&

103

52.3

194

166

o 258.1

164.0

806

968

53

170

21 1

74

606

33

313

1254

57

53

269

143

40

37

2.2

10.0

O. 6

29.7

L44.9

1 .o 167.3

O

2.3

I 13.0

42.1 . L

c -

o. 1

o. 2

i .1.1 J

! 90

Lr 35

E, 2.1

72.4

' r43.6

I

285ì

$18.5

e, o o

247

37.3

51.7

88

.34.2

92.7

23.3

241

4.3

0.6

3.3

15.3

158.0

869

287

4.7

0.6

0.5

13.2

28.7

246

51.8

64.8

' _ i 0.5 . .- . - 29.3

. . . 2.6

G . O . 2

2 . t : 558\

5 22 r!?. 2

17.3

0.51

i.P;2 ~ 2 z . 6

E -E 2 .os -.I u 2 . 2 O

5.c tra\ iec i c

O O O O

-

O. 7

O

O

!73

e3t.j 75.4

29.3

O

24.5 O

-

1H

I 564

31 4

609

196

840

278

709

227

1560

1224

139

21 4

1511

306

O. 4

O

139

51.3

1 . 4

223

514

368.E

-. 82:j 11 .7

190

O

4.E ./ *I,I -- Ji , *bL

740 . *._ &U ?2 2 , 12.2

45.5

2. e O

3. c X3T3I3

1/2H

28 2

362

592

86

367

195

174

226

950

632

87

104

699

562

o. 2

1.1 25.5

55.8

O

123

216.5

278.1

10.7

10.1

O

o. 2

0.3

578

L 3 1 8

11 2

1 6.. O

.

31

SPECIES

OENTEX

ANGOLENSIS

t MACROPHTHALMUS

EPINEPHELUS

. AENEUS

TOTAL

GROUPERS I .

- . . .

. ._ .-

PSEUDUPENEUS 2

1H

606

1261

70.'

97. ! 224

150

5.c

5.:

2.5

O

4.f

47.c

93.;

20. : O

O

8.€

8.C

2. f O

34.5

0.1

1.1 O

3.c

3: 1

I .

38.5

37.3

,'0.9 21: 7

1 2.4

'30.1 i $ i - l t l

j24.a

1 0 7 s

I:b2 l o ~~~~5 117.3

1323

,?à5.5

[e-, 32.3

d3: 2 20: 6

28: 6

O: 5

íC0.6

1/2H 1H

297 44.c

205 528

52.9 196

56.5 205.5

172 44.:

78.5 139.:

5.5 7.c

0.9 16.4

0.4 2.1

2.1 3.2

O O. E

17.5 35.C

3.7 28.9

6.1 2.5

8.8 14.8

4.1 o

22.5 9.3

4.7 17.3

0.4 2.1

2.1 5.9

0.2 27.4

1.2 47.a

O 3.9

18.8 0.9

0.7 8.0

3.9 5.5

68.4 63.2

14.7 3.7

1.3 o c . 5 ; l 24.4

5.0 I 8.6

2.2 109

;7;0 125~7

14;l 110.5

f.32 1 P . c - T

o r .s

I

23) o 1o:o 3525 61: 2

a!:! &3:6

- . ---

1 /2H

19.1

255

58.1

116.1

32.0

77.3

1.5

8.0

3.1

O

2.4

10. o 16.6

O

10.8

o. 9

21.1

21.0

3.1

0.8

O 17.5

1.1

11.2

O

2.1

25.6

4.8

: 0.5

1.8.7

j 1.5

,38.7

113.2

f _ - i c.s

j t .3:

l o

o?, 4

9

rOs3 c42 i: o3

fc?e ---

1H 1/2H

179 58.1

142 105

121.5 71.4

13.6 3.3

18.0 9.5

28.4 13.5

0.9 0.6

22.1 11.4

1.2 o 1.4 0.9

O 0.8

3.0 0.7

2.8 0.3

50.0 30.0

32.3 14.7

0.9 0.6

33.3 '11.4

2.4 2.4

35.0 10.0

94.3 *3.7

1.2 2.3

14.0

8.3 36.3 O \

17.0 9.1

44.6 35.3

'14.4 "11:4

15.0 ;,4:4

i.!.*? 1 7,O

f I

1,5.7 ':O

2.2 i 0.8

i i? L3 'LOC

--._I__

Appendix 11.3 . C . T I xibnsqqa

1H

207

1 .4

13.8

33.1

10.4

229.6

27.3

1.2

2.0

1.8

O. 2

7 . 1

o. 9

8.0

3.2

32.5

1.2

2.0

6.8

1.9

0.3

28.9

20.5

O

O

5.6

O. 4

37.5

' 1.7 . -

. .

1 /2H

116

O

11.2

9.3

4.2

209.7

14.C

O

O

o. 7

0.e O

11.2

O

2.1

16.C

O

1 .2

4.c

o. 7

O

16.6

6.5

8.8

4.1

O

O

s 8.8

: 3.5

, 0.6

i 4.1

1 0

i o i 1 i:: 112.2

5.1

O

i 1 f 5.5

p . 1

f 15.8

I ::: I 4.0

1H

456

87.8

29.6

111.2

89.2

18.8

O. 7

3.3

15.0

0.7

3.9

1.2

O

8.3

23.8

o. 7

5.3

5.5

O. 2

1.4 1.0

2.5

14.8

O

70.9

o. 1

9.1

11.7

11.9

1 .o 1.0

o 5 . JJ:. i i f

1 ! J 3 T i 28.8

36.2

O

5.2

1.0 I..

n42i 5: 52.8 9.9

18.4

11.2

48.4

28.3

*%*$E

1 /2H

364

86.6

O

1040

173.0

5. O

O

2.7

O

1.2

1. i O

14.0

36.3

14.0

O

2.7

6.2

O. 8

o. 9

O. 6

O

11.7

o. 9

33.6

O

13.3

5.5

3.8

1.1

, 0.7

O

! 5.5

27.2

2.6

0 , 1.1

11.5

11.3

3.1

O

9.0

6.3 -

.

1H

6.3

O

0.9

8.0 O

O

17.4

0.9

9.3

1.5

4.1

2.6

10.0

1.5

3.3

52.7

4.5

1.0

2.5

' 3.'0

1.6

2.7

O 1:1

14.0

18.1

3;3

' 1.5

SPECIES 1/2H

O

O. 3

o 2.0

o. 2

1.9

33.9

1.1

2.8

1.4

o 3.0

o 0.9

1.5

2.9

1.0

o

o 6.2

O

2.8

14.4

o 8.0

16.6

0.7

o

UMBRINA

CANARIENSIS

PSEUDOTOLITHUS

SPP.

ZEUS

FABER

RAJA

43i5

- 57:2

1H

16.8 i .

I *

29.6;

1/2H

O

0.2

4.2

4.2

1 .o O

18.4

11.3

4 . 1

O. 7

6.0

3.6

12.3

O. 7

o. 1

O

1.3

o. 4

26.2

O. 2

3.7

6.7

1.2

9.4

4 .3

12.5

9.5

O. 6

3.0

,47.3

1 1:'o

-4 ' 7

nelo

O0

05.

38.0

2.z

11.0 $ . e 8.2

P r s

Qlfl 2.8

O. 6

5?1 Co?$

B2?p

"s"8

$4E6 r14$ -"-

5.4

O. 8

11.7

O

O. 5

O. 4

13.3

6.7

o. 2

1.4

4.2

o. 9

9.2

O

O

1.7

O

O. 4

24 .1

O

1.1

4.0

3.4

6.9

1.0

7.3

5.6

O. 8 +8.4

2:8

O '

j 9:o

i 73

; 4 3 1 - 2

19.3

32.6

f '4:s

4.5

20.1 o

t .o 31:a

4zO

173 O. 7

1.0

c1C7

r OZC

C2?1

8": o ?3?8

3: O - -.

1H 1 /2H

94

2.2

O. 4

3.8

O. 3

9.5

O

9.3

23.4

o. 9

3.4

6.3

1.1

2.9

3.9

13.2

4.9

o. 9

6.5

2.5

4.4

6.2

3.0

3.7

17.5

O

O

t18.8

-,

32.4

9: 5

3.8

89.4 J <:

5 'G c 7 I O

i 530

22i 9

O

2.1 01g2

e597 25?0

54PO [?1!?4

121 6 --_ 1 - L I 1 xibrmqqk Appendix 11.4.

O

O. 5

O

O

O

13.0

O. 7

7.2

3.5

O

1.2

6.0

0.3

1.8

3.6

4. O

4.0

2.6

3.3

O

o. 2

4.8

o. 2

1.5

12.6

10.6

1 0.7

1 8.2 I < z

' C

r6:5

r5.3 f 1 O? r30.1 j 8 . T f P J t3!5

g l i 3 :otr 037

o. 1 21r4

i

1 E o7

b 5 5 4

'25T6

/'i9:6 LI-

1H

2.5

O. 4

O. 8

14.8

9.5

12.3

5.9

1.3

8.0 2.7

2.0

5.0

4.5

2.6

6.3

10.4

14.3

2.5

1.6

3.8

2.0

1.2

9.0

1.4

9.6

21.0

' 0.1

; 2.8

. - . a - 64: O

56: O

t a t 1 6.4

I?

E:*Tf

2325

12 1 EOSb

24'6

9.3

1725

E515 3 07-2 > 5: o : o3 3823

1/2H 1H

O O

O 7.2

1.2 o O o. 9

10.8 5.6

6.5 33.0

4.3 0.9

1.2

8.2 6.8

1.8 4.3

2.6 3.8

0.8 o 0.3 1.3

0.9 9.8

1.0 1.8

6.5 17.4

3.6 13.5

2.3 0.6

2.6 6.8

1.8 11.2

O 1.7

2.0 0.5

2.6 9.9

1.6 10.2

2.5 1.3

2.1 2.2

O ~ 7 6 . 6

2.5 0.7

i12.5 61.0

1 6.1 3.2

6.2 9.8

1-71 8.2

33

SPECIES

Grand t o t a l

Trachurus spp.

Decapterus rhonchus

Scomber japonicus

Boops boops

Brachydeuterus auri tus

3phyraena spp.

lactylopterus vol i tans

'agellus b e l l o t t i i

?parus caeruleostictus

lentex canari ensi s

lentex ango. + macro.

Ipinephelus aeneus

-otal groupers

keudupeneus prayensi I

'riacanthus arenatus

'lectorhynchus medit.

lmbr i na canari ensi s

'seudotolithus spp.

eus faber

aja miraletus

ustelus mustelus

epi

- N - 61

40

14

11

33

18

19

23

40

29

30

19

36

38

34

25

28

20

19

35

43

14

44 -

R/V LOUIS SAUGER

45642

7315

457

106

9276

1607

203

2653

3914

730

278

4466

307

434

628

3710

529

136

106

228

174

438

327

ZX

23513

31 43

216

125

4039

1260

64

1599

2132

426

172

3174

152

272

339

1050

221

26

96

93

129

189

134,

- .A R

1,94

2,33

2,12

0,85

2,30

1,28

3,33

1,66

1,84

1,71

1,62

1,41

2,02

1,60

1,85

3,53

2,39

5,23

1 , l O

2,45

1,35

2,32

2,44

- N - 14

9

5

3

5

10

3

3

12

3

9

10

11

12

7

1

2

7

5

11

13

7

t i -

R/V NDIAGO

7681

1686

150

7.

12

599

10

2

368

24

21

1041

158

162

48

O

65

31

26

79

257

362

1 O0

3834

620

73

O.

3

495

3

0.:

215

11

8

656

77

82

18

1

22

17

16'

51

137

96

58

2,oo

2,72

2,05

1,21

1,71

1,59

2,05

1,98

1.55

1,88

3,77

1,72

o es t imate (2) f o r N p a i r s a l f -hou r catches ( x ) o f th? ach research vessels. R

values 'were n o t computed when N and/or 1 were too smal l .