Rev. bras. oceanogr.. 48(2):131-140. 2000

Aspects of the biology of the Atlantic Midshipman,Porichthys poros issim us (Teleostei, Batrachoididae): an important

by-catch species of shrimp trawling off southern Brazil

Marcelo Vianna '; Acácio R. G. Tomas I & José R. Verani 2

'Instituto de Pesca

(Av. Bartolomeu de Gusmão, 192, 11030-906 Santos, SP, Brasil)

2Universidade Federal de São Carlos

(Caixa Postal 676, 13565-905 São Carlos. SP, Brasil)

. Abstract: Trawl fishing for pink shrimp (Fmfantepenaeus brasiliensis and F. paulensis)catches large quantities of by-cateh fish, discarded due to their having no commercial value.As these species have rarely been studied, the impact of fisheries on these populations is notknown. This contribution studies the biology of a species of no commercial value, the Atlanticmidshipman Porichthys porosissimus. The last haul fToma commercial trawler, operating onthe northern coast of São Paulo State and the southern coast of Rio de Janeiro State, waspreserved (from being discarded) and identified at the species leveI. It was observed that thisfishery affects the juvenile population which is intluenced by the seasonal variation of thewater masses. Growth parameters were estimated: Lx = 37.0 em, K = 0.285 year,l.Instantaneous mortality coefficients were estimated: Z=2.14. M=0.63, F= 1.51, S= 11.8 andE=0.71. Considering the intensive activity ofthe shrimp trawl tleet operating in this area, thedeleterious action of trawling is considered as of high impact. The stock managementmeasures applied for pink shrimp are without effect regarding P. porosissimus, which has itsspawning period before the c\osed season and its recruitment peak after it. The resu1ts showoverfishing and the need to apply measures to reduce trawling action, such as adequatepolicies, introduction of selectivity devices and the creation of exc\usion zones for trawlfishing.

. Resumo: A pesca de arrasto para a captura do camarão-rosa (Farfàntepenaeus brasiliensis eF. paulensis) captura grande quantidade de peixes considerados fauna acompanhante que sãodescartados por não possuírem valor comercial. Assim, pouco aparecem em trabalhos debiologia pesqueira com dados de desembarque, sendo a ação pesqueira sobre estas populaçõespouco conhecida. Neste estudo, um arrasto mensal de um barco de pesca foi desembarcadosem que nada do material capturado fosse descartado. Analisou-se Porichthys porosissimus doqual foram avaliados aspectos biológicos e pesqueiros. Estimou-se, para a espécie: Lx = 37,0cm, K = 0,285 ano", Rn=180 (xIO"), Z=2,14, M=0,63, F=1,51, S=II,8 e E=0,71. Embora nãointencional, a pesca age principalmente em jovens e ocorre durante o ano todo, sendointluenciada pela sazonalidade das massas d'água. Considerando a fTota operante na área, oefeito do arrasto é considerável, já que a taxa de aproveitamento é nula. As medidas deproteção de estoque aplicadas ao camarão-rosa não são efetivas para P. poro.l'is.l'imus,queapresenta o período de desova anterior à época do defeso e o pico de recrutamento pesqueiroposterior. Os resultados indicam sobrepesca da população estudada sendo necessário medidasde modo a diminuir a ação deletéria do arrasto, como a adequação da legislação, uso dedispositivos de seletividade e a criação de zonas de exclusão de arrasto.

. Descriptors: Porichthys porosissimus, By-catch, Shrimp trawls, Mortality: São Paulo coast.

. Descritores: Porichthys porosissimus, Fauna acompanhante, Pesca de arrasto de camarão,Mortalidade: Litoral de São Paulo.

132 Rev. bras. oceanogr., 48(2). 2000

Introduction

The south-eastern region has one of the widestcontinental shelves on the Brazilian coast, which allowsthe trawl fisheries of severaI shrimp and groundfishspecies. Most of the fishing boats operate between 4and 90 meter depths. Due to the low selectivity of thetrawlers and the high diversity of catches, shrimptrawlers are considered responsible for the greater partof the by-catch. By definition, by-catch comprises anyindividual, of any size or species, caught together withthe target species. The species composition of the by-catch has some interaction with the target species, e.g.,the shared spatial distribution. Only a portion of thetotal catch is landed and sold (either as a singlecategory, or grouped by similarity or in a groups ofseveral species). The others are discarded because theyhave no commercial value. They are composed ofimmature and/or smalI individuais (Alverson et aI.,1994) and are not included in the catch statistics, whichdiminishes the efficiency ofthe catch control (Coelho eta!., 1986), and is harmful to several species andpopulation strata (Haimovici & Mendonça, 1996).

Among shrimp trawl fisheries, finfish are themost abundant and of economic importance (Coelho eta!., op. cit.). Despite the large number ofspecies, few ofthem constitute a large biomass or number ofindividuais. Though this is welI known, the biology offish species along the northern coast of São Paulo Stateis further harmed. Regional studies on the biologicaldata of species fished wilI contribute to the formulationof fishery management plans, including fisherypolicies. The updating and monitoring of thisinformation on the by-catch species will allow someestimate of the population parameters to be made, suchas the von BertalanffY growth curve, as also of theirreproductive activities. For Haimovici & Macieira(\ 981), quantifYing effects of discarding on fish stocksmay be evaluated by growth studies, mortality,distribution and abundance ofthose species.

The study area has already been investigatedintensively by the Oceanographic Institute of theUniversity of São Paulo (Rossi-Wongtschowski &Paes, 1993). The opportunity to compare the catchcomposition obtained ITomscientific investigation withcommercial catches over a period of 10 years, isconvenient for checking occasional oscilIations of thepopulation parameters of the groundfish species.Castro-Filho et a!. (\ 987) described the hydrodynamiccomplexity of the region, showing that the cold watermass (South Atlantic Central Water) has a decisive rolein the biological productivity of the coastal regionduring the summer season.

This contribution shows some aspects of thefishery biology of the Atlantic midshipman, Porichthysporosissimus (Valenciennes, 1837) (Teleostei,

Batrachoididae), to show the catch composition, growthand mortality estimates, as welI as reproductiveactivities and the impact of the fisheries on thepopulation sampled.

Material and methods

The Atlantic midshipman, Porichthysporosissimus, of the region was formerly studied byPeria (\ 995) using data obtained ITomscientific cruisesconducted during 1984/85. The biological samplingwas made once a month over the 14-month period ITomJuly 1995 to August 1996 on commercial shrimptrawlers. The first two months were used to adapt themethodology and so these data were not used toestimate the biological parameters. Because the by-catch information of shrimp trawling in the region waslimited to the coastal region along the 15 m isobath, thesamples were obtained through catches ITomover 30 mdeep of the pink shrimp (Farfantepenaeus brasiliensisand F paulensis) fisheries. Considering the highmobility of the fishing boats, the catch stations wererepresented by the nearest geographical placecorresponding to a central point ofthe trawling area.

Data were provided by a haul made IToma 17m double-outrigger otter trawl boat on the northerncoast of São Paulo State andlor southern coast of Rio de

Janeiro State (only one of the two nets was analysed).The dimensions of the net used were 25 mm mesh size

ITomopposite knots on codend; length of 18m on theupper side of the net and 20 m on the bottom. Thetowing speed was maintained around 2.5 knots andeach haul lasted between two and three hours (Tab. I).The entire catch was landed (no discarding was done).To avoid losing data during the 90-day closed seasonfor pink shrimp fisheries (February 15thto May 15th),samples were obtained with a sea-bob shrimp(Xiphopenaeus kroyeri) fishery net (9 m in length, 6 mon the bottom side of the net and 15 mm mesh size on

codend), operating at an average depth of 23 m.Although there are some differences in design andmesh size among nets employed for pink and sea-bobshrimps, it is supposed that after some time ofoperation their selectivity (i.e., retained size ofindividuaIs) may be considered similar. No furtherconsiderations regarding fish escaping due to differentshapes (and, consequently, tow ratios) of the nets areraised here.

Ali landings were done at Ubatuba city.Data of the fishing area, depth and duration of haulwere registered by interviewing the masterfishermen. Catches were separated and weighed firstas a whole (by species) and then individually. Eachspecimen had its total length (in 0.5 cm) and totalweight (in 1 g. after drying in paper) registered. Totalnumber (N), total biomass and length trequency were

VIANNA et aI.: Fishery and biology of Porichthys porosissimus 133

also obtained. The relative numerical abundance(RNA) was based on the number of individuaIs perhour of haul (ind/h). Catch-per-unit-of-effort (CPUE)was obtained by the ratio of total weight (in grams)per hour of haul (g/h). CPUE values must be seen asexclusively for purposes of comparison between thetarget species and P. porosissimus. AlI data weregrouped bimonthly. The composition and populationstructure were obtained by analysing the number ofindividuaIs per length class for the sex groupings.

Several methods were used for the estimates

of growth parameters. Taylor's expression (Pauly &Martosubroto, 1980), Loo = L max / 0.95 was used,where Lmax was obtained through data gathered inthis study, from scientific literature and unpublisheddata furnished by other colIeagues in the same area.Then Wetheral1's method (1986, as modified byPauly, 1986), available in FlSAT package (FAO-ICLARM Stock Assessment Tools), and the ELEFAN Iroutine (Pauly & David, 1981) were used for dataanalysis. Considering, first, the option for oscillatorygrowth, the estimates of the oscillation factor (C) andthe winter point (WP) were made, based on theseasonal infIuence of South Atlantic Central Water(SACW) in the area and on the seasonaltemperature variability which varies (5.5°C, Castro-Filho et aI., 1987, or 4.5°C, Rocha, 1990), as betweenwinter and summer, thus obtaining C estimates ofbetween 0.45 and 0.55. Taking December as the coldestwater month (due to SACW), WP was estimated at0.91. AlI samples (except those for March andApril) were grouped bimonthly and the filesgenerated by the smoothing routine were refined byrunning average. Values of Loo and K were chosen,assuming the goodness of fit index (Rn) withbiological significance for the species, considered asindicative of the estimate of Loo obtained by formermethods. For the same Rn values, the highest values

Table I. Generaldataof tisheriessamplesto this studyin southernBrazil.

for K and the lowest for Loowere chosen, consideringthat the method overestimates Loo (and consequentlyunderestimates K) (Isaac-Nahum, 1989).

The growth performance index $' (Munro &Pauly, 1983) ($'= log K + 2 log Loo)was applied to Kand Loovalues estimated by ELEFAN I. If differencesoccurred in $' values, they were rejected (Sparre &Venema, 1992). The estimates of Looand K were alsomade taking non oscilIatory growth (C and WP equal tozero) into consideration.

The length-weight relationship (LWR) wascalculated by the power equation (Wt = a L/), afterthe log-transformation of Wt and Lb and using theleast square method (Sparre & Venema, 1992). The 1-test (Zar, 1984) was applied to verifYthe hypothesis ofb = 3.0 (HO), which defines the type of growth (asalometric or isometric).

The relative condition factor (Kn) wascalculated (Kn = Wt / We) by using expected weight(We) (LeCren, 195"1)for a certain length (We = a Lth),alIowing the comparison of statistical values of Knwith 1.0. Kn values were estimated individualIy(excluding juveniles), and bimonthly averages and theirrespective confidence intervals (CI; a = 0.05) obtained.The cut-off length attributed to maturity for P.porosissimus was 15.5 cm, based on the gonadalobservations. The t-test (Zar, 1984) was applied toverifYif values of bimonthly averages of Kn were equalto 1.0.

Fishery recruitment was estimated graphicalIyby length class frequency distributions. Therelationship between target species (pink shrimp)catches and midshipman catches was evaluated usingbimonthly values of CPUE for each one. A1though thetrawl nets for pink and sea-bob shrimps had differentdesigns, after some minutes of trawling their selectivitymay be considerably similar.

Sample Coordinates Month Depth Duration of

. (m). trawl (h)

23°48'S 145°0 I'W July/9545 02:00 5.0

I! 23°40'S 1 44°53'W Augi95 43 03:00 4.0

III 23°41 's 1 45°02'W Sept/95 4] 01:00 5.0

IV 23°36'S 145°01 'W Oct/95 37 02:00 4.0

V 23°21'S 144°32'W Nov/95 35 03:30 4.0

VI 23°38'S 145°04'W Oec/95 35 01:30 4.0

VI! 23"28'S 144°55'W Jan/96 31 03:30 3.5

VII! 23"27'S 144°45'W Feb/96 37 02:30 4.0

IX 23"25'S 144°52'W Mar/96 22 11:00 3.0

X 23"26'S 144°52'W Apr/96 24 10:00 2.0

XI 23"26'S 144°35'W May/96 51 02:00 4.0

XI! 23°42'S 145°06'W Jooe/96 35 01:30 4.0

XlII 23°38'S 145°0 I'W July/96 42 01:00 6.0

XIV 23"25'S 1 44°40'W Augi96 41 02:30 4.0

134 Rev. bras. oceanogr., 48(2), 2000

The total mortality (Z) estimates were madeusing the equations of Beverton & Holt (1956) (Z = K[(Loo - L')/(Lm - L')]) and of Ssentengo & Larkin(1973) (Z = (n K)/{(n + 1) ln [(Loo- L')/(Loo- Lm)]}),where L' is the length at which 100% of individuaIs arevulnerable to fishery and Lm the average length of allindividuaIs larger than L'. The catch curve was alsoobtained using the FISAT programo Natural mortality(M) was estimated by Pauly's empirical equation (log M= 0.0066 - 0.279 log Loo+ 0.6543 log K + 0.4634 logT) (Pauly, 1980), where habitat temperature (T) wasconsidered to be 18.2 oCOThe final estimate for Z wasadmitted as the average of all estimated values and thenapplied to obtain an estimate for fishery mortality (F),stock exploitation ratio (E) and survivaI rate (S) usingthe following equations: Z = F + M; E = FIZ; S = e -z.

Results

Analysis of data was made based on the 2,897individuaIs of Porichthys porosissimus, :fi-omall thesamples collected over the 14-month period with abiomass of 99.6 kg. The highest values of CPUE andRNA (Fig. 1) were registered for November-Decemberand the lowest in July-August. Due to the pink shrimpclosed season, the values for March-April were nottaken into consideration

9000

20

. --. 1808000 ~CPUE ~ 160

RNA7000

140 Ê,;

120 :!..,;

-'-100 §..

-->- 80 ~~

60 ~40 &

6000

..; 5000:>~ 4000

3000

2000

1000

osepoct mayJJn

ojJlaugnov dec jan feb rrarapr

bimonthlyperiod

Fig. 1. Bimonthly catch per unit of effort (CPUE glh) andreiative abundance in number (ind./h) to Porichthysporosissimus in southern Brazil.

The pink shrimp presented the highest CPUEin July-August and the Iowest in January-February,remaining constant the period :fi-omMay to December(average close to 1,600 gIh) (Fig. 2).

The Iength range of the Atlantic midshipmanwas :fi-om 2.5 to 29.0 cm. By analysing thebimonthly length :fi-equencydistribution (Fig. 3), the16.0 em class may be considered as the recruitmentclass, beginning in May-June and attaining its peak inJuly-August. In September-October the smallindividuaIs were present in large numbers, indicating a

possible continued recruitment as:fi-om July-August(Fig. 4).

'1200i~ 1000

~."

j.mfeb jutaug

bímonthlyperiod

Fig. 2. Bimonthly catch per unit of effort (CPUE g/h) to thepink shrimp (Farfantepenaeus brasiliensis and Fpaulensis) in southern Brazil.

2:i

60% (

I!

40% ..!;!

i'

345670%

8 9 10 11 12 13 14 15 16 17 18 H.20 21 22 23 24 25 26 27 28 29

length(cm)

Fig. 3. Distribution of classes oftotallength &equenciesofPorichthys porosissimus in southern Brazil.

The largest specimen registered in theliterature measured 32.0 cm (Figueiredo & Menezes,1978) and this value was considered as Lmax,generating an estimated Loo of 33.7 cm. Theestimate for Looby Wetheral1's method was 33.37 em(r = -0.964). The ELEFAN I estimates for growthparameters are presented on Table 2. These valuesindicated six age classes (Fig. 5). Estimates :fi-omnonoscillatory were smaller than those :fi-omoscillatorygrowth. The ELEFAN I values were higher than thosegiven by other methods, but the tendency tounderestimate Loo values deserves mention (Isaac-Nahum, 1989).

The length-weight relationship of theAtlantic midshipman was expressed by theequation: Wt = 0.00498 Lt 3.148(r2 = 0.941) and the t-test for parameter b confirms the growth withpositive alometry (calculated t = 8.398 > critical t, p <0.05). The reIative condition factor (Kn) for thespecies presented values higher than 1.0,considering significance tests for three bimonthlyperiods (September/October, January/February,July/August) (Fig. 6 and Tab. 3). Instantaneousmortality coefficients, exploitation rate and survivalrate are shown in Table 4.

-----------

"-...." ""

".

VIANNA el aI.: Fishery and biology 01'Porichlhys po/"Osissimlls 135

20

nov - de c

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

30

3 o25

~o~1 5..~1 ou. 5

o

ja n - fe b

40

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

20

10

...

.JO

.J1996

F

mar - apr

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

J1995

5 D

ma y - ju n

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

o N

jul-aug

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

A

Fig. 4. Bimonth distribution 01'classes 01'total kngth trequencies 01'Porichlhys porosissimlls in southern Brazil.

M

Fig. 5. Distribution 01' classes 01' total length restructured fi.equencies and seasonal growth curves 01' Porichlhys

porosissimus (by ELEFAN I) in southern Brazil.

3025

>. 20u

I 1510

.:: 5o

25

20

i5.ia

136 Rev. bras. oceanogr., 48(2). 2000

Table 2. Growth parameter estimates by ELEFAN 1routinein Fisat for Porichthys porosissimus in southernBrazil.

The length-weight relationship of the Atlanticmidshipman was expressed by the equation: Wt =0.00498 L/148 (r2 = 0.941) and the t-test for parameterb confirms the growth with positive alometry(calculated t = 8.398 > critical t, p < 0.05). The relativecondition factor (Kn) for the species presented valueshigher than 1.0, considering significance tests for threebimonthly periods (September/October,January/February, July/August) (Fig. 6 and Tab. 3).Instantaneous mortality coefficients, exploitation rateand survival rate are shown in Table 4.

.li

,0,60 i (n:::191) (n=341) (n=51)(n=322) (n=3)(n=1B4)

sep ocU95 novdec jan febl96 mar apr8imonth

maYJun jul aug

Fig. 6. Bimonth averages values ofrelative condition factor(Kn) and confidence intervals (Cl) of Porichthysporosissimus. (excluding juveniles) in southernBrazil.

The ratios of the pink shrimp catches: Atlanticmidshipman catches (Tab. 5) show higher catches forthis latter species. Otherwise, these ratios registeredconsiderable variation over the study period, due to theseasonal variation in the abundance of the Atlanticmidshipman.

Discussion

Porichthys porosissimus had high fi-equenciesof occurrence throughout the year (Haimovici et aI.,1996; Rocha, 1990; Natali-Neto, 1994). A relationshipbetween the species and cold waters masses in the 20-50 m depth stratum, occupying the area dominated bySouth Atlantic Central Water and the transitional areawith Coastal Water, was confirmed by Natali-Neto(1994). Along the Rio Grande do Sul coast (28° - 33°S)the species has been reported in waters deeper than26 m (Haimovici & Habiaga, 1982; Haimovicl et aI.,1996). Due to this depth distribution model it may besuggested that P. porosissimus is of structuralimportance for the Ubatuba region and is related to thebottom type, having wide distribution in the area withno water mass dependence (Rossi-Wongtschowski &Paes, 1993; Peria, 1995), contradicting Natali-Neto (op.cit.) for whom P. porosissimus occupies areas with coldwaters under the direct influence of SACW. This mayexplain the deep strata occupied by this species andmay also be used to explaining the variation inabundancerecordedbyRocha(1990)- highervaluesinsummer - and by Natali-Neto (op. cit.) - higherabundance in winter. 1n this study, the highest catchesoccurred at the end of spring and the lowest catches inwinter, suggesting some relationship with SACW.

Table 3. Average values of relative condition factor (Kn). confidence intervals (Cl), standarddeviations (Sd) and t-tests for Porichthys porosissimus (excluding juveniJes) insouthern Brazil.

Table 4. Bimonthly relationship between pink shrimp and Porichthys porosissimus (in g) andgeneral average (not including March-ApriJ) in southern Brazil.

Jan-Feb0.49:1

!\1ar~Apr May-Jun . _Jul~,'\lIg. . Average1

L::c(cm) K(year") C WP Rn(x 10') cJ>'

370 0.285 0.3 0.91 180 2.83

37.0 0.28 171 2.83

1,40 lI

U I !!! !

't100 II !

Kn CI Sd N t calculate t critic results

Sep-Oct 1.062 0.016 0.112 191 7.648 1.653 Kn > 1.0*

Nov-Dec 1.0 II 0.014 0.129 322 1.546 1.650 Kn = 1.0

Jan-Feb 1.066 0.024 0.165 184 5.391 1.653 Kn> 1.0*

Mar-Apr 1.159 0.143 0.126 3 2.184 2.900 Kn = 1.0

May-Jun 0.930 0.015 0.136 341 9.511 1.650 Kn < 1.0

Jul-Aug 1.099 0.032 0.115 51 6.156

VIANNA et aI.: Fishery and biology of Porichthys porosissimus 137

Table 5. Comparison ofmortalities ratios estimates: total (Z). natural (M), and fishing (F); and survival (S) amongthe present study and Peria's (1995) in southern Brazil.

Taking the pink shrimp as the target species, itis to be expected that fisheries should be undertaken inthose areas in which it is most abundance.

The length composition and recruitment analysisare important for the characterization of the populationstructure. Peria (1995) studied the population dynamicsof P. porosissimus off the South Brazilian coast andpresented similar findings to those of the presentcontribution. The recruitment period was confirmed inwinter, although the second recruitment period showedby that author in April was not observed in this study.Haimovici et ai. (1996) suggested that the recruitmentfor the species was observed throughout year in the RioGrande do Sul state, but not in this region. As shownby length composition, different population strataoccupied the sampled area with the recruitmentoccurring in small batches during the year, thoughpeaks may be seen in the early summer and winter.

The ELEFAN I and Wetherall methods are thebest indicators for tropical species with low life cycleand quick growth (Isaac-Nahum, 1989), but theinterpretation ofthe results depends on the sample sizeand the biological information regarding the species.Problems in estimating growth parameters employinglength class distribution provided by commercialfishery landing data are due to a lack of information onthe discarded portion of the specimens (Sparre &Venema,1992).

Analysing the results, the oscillatory growthpattern was considered to be closer to reality. ThisassÜmption differs fTom Peria's (1995) choice. Sparre& Venema (1992) affirmed that oceanographic changesin tropical areas are reflected in seasonality of growthmaking it oscillatory. Avila-Silva (1996) attributes tothe penetration of the SACW onto in the continentalshelf the ring formation of the urohyal and scales ofPagrus pagrus of southeastern Brazil, reflected in theWP values which thus corroborate the methodsadopted. On the other hand, Isaac-Nahum (1989)demonstrated that when oscillatory growth is not takeninto consideration there is a resulting low variationamong estimated values, as is also to be observed inthis study (Tab. 4). The adoption of oscillatory growthshows better values for the fit of goodness index (Rn).The best fit of goodness index was obtained by keepingthe winter point (WP) earlier, which reduced the

amplitude of oscillation (C) to 0.3, indicating a lowervariation in water temperature as between summer andwinter. This behaviour makes sense when P.

porosissimus is considered a subtropical species relatedto cooler water and as thus lessen influence of CoastalWater. Peria (1995) estimated seven (ELEFAN 1) andsix (Bhattacharya) age classes (attributed to the lowestnumber of sampled age classes of older specimens).The estimate for <1>:shows Loo and K relationships asvalid, with no discrepancies in the values registered,and similar to those estimated by Peria (op. cil.).

The Atlantic midshipman shows a positivealometric growth, indicating a better application to therelative condition factor (Kn).

The Atlantic midshipman shows two periodsofhigh Kn values (winter and middle of summer). Thismay be indicative of partial spawning, but the peakscoincided with recruitment and the high value observedin summer with intense gonadal development.DeMartini (1990) registered a reproductive migrationin early summer for the co-generic species P. notatus.The same may be suggested for P. porosissimus as thesummer as the most intense reproductive period. If Knvalues higher than 1.0 are taken to represent the periodof highest gonadal development, the species spawnsbetween November and February, the period with thelargest food offer, high adult concentration and warmercoastal water. This spawning pattern was confirmed byKatsuragawa et ai. (1993) who demonstrated the highegg abundance of several groundfish species in theUbatuba region during the austral summer. The highestvolume of ichthyoplankton in summer coincides withthe SACW penetration into the coastal region,enhancing the availability of food for fish larvae. Asthis period is different fTomthe 90-day closed fishery

ZB&H ZS&L Z c.cath Z average M Pauly F S E References

1.47 1.60 1.54 1.53 0.59 0.95 21.6 0.62 Peria

2.29 2.46 l.75 2.17 0.62 1.55 11.4 0.71 n. o.

2.33 2.45 1.64 2.14 0.63 1.51 11.8 0.71 o.

Obs.: Beverton & Holt (B & H). Ssentongo & Larkin (S & L) and catch curve (c. catch), non oscillatory (n. o.); oscillatory (o.).

Table 6. Values of growth estimates for Porichthys

porosissimus in southern Brazil.

Loo(cm) K (year'l) Methods Rn(IO") References

36.9 0270 Elefan I n. o. 115 Pelia (1995)

37.0 0.280 Elefan I n. o. 171 this paper

37.0 0.285 EIefan I o. 180 tms paper

138 Rev. bras. oceanogr., 48(2). 2000

season for shrimp between February and May, thereproductive behaviour suggests that commercialfisheries were having an impact on mature specimensof P. porosissimus.

The Atlantic midshipman, a species endemic tothe Argentinian Zoogeographical Marine Province(Figueiredo, 1981), is replaced in the Caribbean by P.plectrodon, this latter is used as human food (Cervigon,et aI., 1992). Figueiredo (op. cit.) reported only a fewmorphological differences between P. porosissimusand P. plectrodon and, for this reason, they areconsidered by some authors to be synonymous.This similarity suggests that discards are cultural, aspointed out by Haimovici & Mendonça (1996). ln SãoPaulo State, the species is one of the most abundant inthe by-catch of the sea-bob shrimp fishery (Paiva-Filho& Schmiegelow, 1986; Graça-Lopes, 1996) but they areentirely discarded. The lack of commercial value masksthe high catches of the species in shrimp fishery.Vianna (1998) observed that P. porosissimus is thespecies with highest CPUE and relative numericalabundance. To Alverson et aI. (1994), mortalityestimates by discards are of importance because theyfurnish additional data on fishery mortality beyond thatprovided by the official landings. Total mortalities ofthe species obtained by Peria (1995) were higher thanthose ofthe present study, but as natural mortality mustbe quite similar, fisherymortality must be responsiblefor that increase. Avila-Silva (1996) interpreted theabrupt decrease in the abundance of the greater lengthclasses as possibly due to fishery as mortality, but thereis also a possibility that it may be related to their lesservulnerability to fishing because of the seasonalmigration of some individuaIs. This pattem wasobserved in the present study and probably themortalities have been overestimated.

Vianna (1998) has observed an increase in thepink shrimp fleet over recent decades in the Ubatubaregion. Values of exploitation rate over 0.5 indicateoverexploitation (Pauly, 1983), thus demonstrating thedeleterious fishery action on the population studied.Alverson et aI. (1994) confirmed the impact caused bynon intentional catches and this has different effects oneach species in terms of reduced reproductive ratio,high parental aid and low natural mortality. As farsmall fish, due to their high mortality, other factorsmay impact growth and recruitment, resulting insignificant losses in the by-catch population.

Discards of by-catch are of great importance, aspointed out in Agenda 21 (UNCED, 1992) and recentworld-wide efforts have been made to monitor andreduce them (Alverson et aI., 1994). There are manyways of reducing discards of Porichthys porosissimusand others fish, through stimulating the use of theproduction of by-products, adequate fishery policies,

introduction of selectivity devices and the creation ofexclusion zones for trawl fishing.

Acknowledgements

The authors thanks Drs. Alberto Peret andNivaldo Nordi (UFSCar); Erica Caramaschi (UFRJ);Francisco M.S. Braga (Unesp) and Carmem Rossi-Wongtschowski (IOUSP) by criticisms to the PhDthesis of the former author that originates thiscontribution. CAPES and CNPq for the scholarships toM.V., and two anonymous referees for useful commentsto previous versions ofthe text.

References

Alverson, D. L.; Freeberg, M. H.; Pope, J. G. &Murawski, S. A 1994. A global assessment offisheries by-catch and discards. FAO Fish. Tech.Pap., 339:1-233.

Avila-Silva, A O. da 1996. Idade, crescimento,mortalidade e aspectos reprodutivos do Pargo,Pagrus pagrus (Teleostei: Sparidae), na costa doEstado de São Paulo e adjacências. Dissertação demestrado. Universidade de São Paulo, InstitutoOceanográfico. 116p.

Beverton, R. J. H. & Holt, S. J. 1956. A review ofmethods for estimating mortality rates in exploitedfish populations, with special reference resourcesof bias in catch sampling. Rapp. P-v. Reún. Cons.Perm. Int. Explor. Mer, 140:67-83.

Castro Filho, B. M.; Miranda, L. B. de & Miyao, S. Y.1987. Condições hidrográficas na plataformacontinental ao largo de Ubatuba: variaçõessazonais e em média escala. Bolm Inst. oceanogr.,S Paulo, 35(2):135-151.

Cervigon, F.; Cipriani, R. & Fischer, W. 1992. Guia decampo de Ias espécies comerciales marinas y deáguas salobres de Ia costa septentrional de SurAmérica. Roma, FAO. 513p.

Coelho, J. AP.; Puzzi, A; Graça-Lopes, R.;Rodrigues, E. S. & Prieto Jr., O. 1986. Análise darejeição de peixes na pesca artesanal dirigida aocamarão sete-barbas (Xiphopenaeus kroyeri) nolitoral do Estado de São Paulo. B. Inst. Pesca, SãoPaulo, 13(2):51-61.

DeMartini, E. E. 1990. Annual variations in fecundityegg size and condition of the plainfin midshipman(Porichthys notatus). Copéia, (3):850-855.

VIANNA et aI.: Fishery and biology of Porichthys porosissimus 139

Figueiredo, 1. L. & Menezes, N. A. 1978. Manual depeixes marinhos do sudeste do Brasil. 11.Teleostei(1). Universidade de São Paulo, Museu deZoologia. 110p.

Figueiredo, 1. L. 1981. Estudo das distribuiçõesendêmicas de peixes da província zoogeográficamarinha argentina. Tese de doutorado.Universidade de São Paulo, Instituto deBiociências. 121p.

Graça-Lopes, R. 1996. A pesca do camarão sete-barbas Xiphopenaeus kroyeri Heller (1862) e suafauna acompanhante no litoral do Estado de SãoPaulo. Tese de doutorado. Universidade EstadualPaulista, Instituto de Biociências de Rio Claro.96p.

Haimovici, M. & Maceira, R. P. 1981. Observaçõessobre seleção a bordo e rejeição na pesca dearrasto de fundo no Rio Grande do Sul. In:CONGRESSO BRASILEIRO DE ENGENHARIADE PESCA 11.Recife, 1981. Anais. Recife, AEP.pAOI-412.

Haimovici, M. & Habiaga, R. P. 1982. Rejeição a bordona pesca de arrasto de fundo no litoral do RioGrande do Sul num cruzeiro de primavera. Doc.Tecn. FURG, 2:1-14.

Haimovici, M.; Martins, A. S. & Vieira, P. C. 1996.Distribuição e abundância de peixes teleósteosdemersais sobre a plataforma continental do sul doBrasil. Rev. Brasil. Biol., 56 (1):27-50.

Haimovici, M. & Mendonça, 1. T. 1996. Descartes dafauna acompanhante na pesca de arrasto detangones dirigida a linguados e camarões naplataforma continental do sul do Brasil. Atlântica,Rio Grande, 18:161-177.

Isaac-Nahum, V. 1. 1989. Analysis of methods for theestimation of fish growth parameters. based ondata 1Tomthe family Sciaenidae and on simulateddata. Tese de doutorado. Universitat Zu Kiel,Fakultat der Christian-Albrechts. 244p.

Katsuragawa, M.; Matsuura, Y.; Suzuki, K.; Dias, 1. F.& Spach, H. L. 1993. O ictioplancton ao largo deUbatuba. SP: composição, distribuição eocorrência sazonal (1985-1988). Publção esp. Inst.oceanogr., S Paulo, (10): 85-121.

LeCren, E. D. 1951. The length-weight relationshipand seasonal cyde in gonad weight and conditionin perch Perca fluviatilis. 1. Anim. Ecol.,20(2):201-219.

Munro, 1. L. & Pauly, D. 1983. A simple method forcomparing the growth of fishes and invertebrates.Fishbyte, I (1):5-6.

Natali-Neto, J. F. 1994. Ictiofauna do ecossistematropical marinho de Ubatuba(23°36' - 24°22' S;44°33' - 45°08 I W). SP - Brasil, entre 50 e 100metros de profundidade: composição, distribuição,abundância e diversidade. Dissertação demestrado. Universidade de São Paulo, InstitutoOceanográfico.2v.

Paiva-Filho, A. M. & Schmiegelow, 1. M. M. 1986.Estudo sobre a ictiofauna acompanhante da pescado camarão sete-barbas (Xiphopenaeus kroyeri)nas proximidades da Baía de Santos-SP. 1-Aspectos quantitativos. Bolm Insl. oceanogr., SPaulo, 34:79-85.

Pauly, D. 1980. On the interrelationships betweennatural mortality. growth parameters and meanenvironmental temperature in the 175 fishstocks. 1. Cons. Inl. Explor. Mer, 39(2):175-192.

Pauly, D. & Martosubroto, P. 1980. The populationdynamics of Nemipterus marginatus off WestemKalimantan, South China Sea. J. Fish Biol.,17(3):263-273.

Pauly, D. & David, N. 1981. ELEFAN I, a BASICprogram for the objective extraction of growthparameters from length-1Tequency data.Meeresforschung, 28(4):205-211.

Pauly, D. 1983. Some simple methods for theassessment of tropical fish stocks. FAO Fish.Tech. Pap., 234: 52p.

Pauly, D. 1986. On improving operation and use oftheElefan programs. Parl. 11. Improving theestimation ofL<X:J.Fishbyte, 4(1):18-20.

Peria, C. G. 1995. Estimativa da taxa de produção eda relação produção-biomassa média (P-B) depeixes demersais do ecossistema costeiro deUbatuba. SP. Brasil. Dissertação de mestrado.Universidade de São Paulo, InstitutoOceanográfico. 141p.

Rocha, G. R. A. 1990. Distribuição, abundância ediversidade da ictiofauna na região deUbatuba-SP (23°20'S - 24°00'S; 44°30'W -

45°30'W), Brasil. Dissertação de mestrado.Universidade de São Paulo, InstitutoOceanográfico. 2v.

140 Rev. bras. oceanogr.. 48(2). 2000

Rossi-Wongtschowski, C. L. D. B. & Paes, E. T.1993. Padrões espaciais e temporais dacomunidade de peixes demersais do litoral nortedo Estado de São Paulo-Ubatuba. Brasil.

Publção. esp. Inst. oceanogr., S Paulo, (10):169-188.

Sparre, P. & Venema, S. C. 1992. Introduction totropical fish stock assessment. Parto I. Manual.FAO Fish. Tech. Pap., 306: 376p.

Ssentongo, G. W. & Larkin, P. A. 1973. Somesimple methods of estimating mortality rates ofexploited fish populations. 1. Fish. Res. Bd.Can., 30(5):695-698.

UNCED. United Nations Conference onEnvironment and Development. 1992. Agenda21. Capo 17.

Vianna, M. 1998. Análise de populações de peixesteleósteos acompanhantes da pesca de arrastodo camarão-rosa (Penaeus brasiliensis e P.paulensis), em Ubatuba, SP: captura.crescimento e mortalidade. Tese de doutorado.Universidade Federal de São Carlos, PPG-ERN.114p.

WetheraIl, 1. A. 1986. A new method for estimatinggrowth and mortality parameters fTom length -fTequency data. Fish byte, 4 (I): 12-14.

Zar, J. H. 1984. Biostatistical analysis. EnglewoodCliffs, Prentice-Hall. 718p.

(Manuscript received 16 September 1999; revised24 April 2000; accepted 07 December 2000)