arquip~lagq - Repositório da Universidade dos Açores

AGE AND
AND GROWfH
GROWTH OF
OF THE
THE RED
RED PORGY,
PORGY, PAGRUS
PAGRUS PAGRUS
PAGRUS (LINNAEUS,
(LINNAEUS,
AGE
1758)(PISCES:
(PISCES :SPARIDAE)
SPARIDAE)lN
INAZOREAN
AZOREANWATERS
WATERS
1758)
SERAFIM&&HELENA
HELENAM.
M.KRUG
KRUG
MARIA
MARIAPAULA
P A U L AP.PSERAFIM
SERAFIM,
MARIA
PAULA
HELENA
M.KRUG
KRUG1995.
1995.Age
Ageand
andgrowth
growthofofthe
thered
red
SERAFIM,
MARIA
PAULA
P.P.&&HELENA
M.
ARQUIP~LAGQ
ARQ
UIPÉLAGO porgy,
porgy, Pagrus
Pagruspagrus
pagrus (Linnaeus,
(Linnaeus, 1758)
1758)(Pisces:
(Pisces: Sparidae),
Sparidae),inin Azorean
Azorean waters.
waters.
Arquipe'lago.
Lifeand
andMarine
MarineSciences
Sciences 13A:11-20.
13A:ll-20.Angra
Angrado
d oHeroísmo.
Heroismo. ISSN
ISSN
Arquipélago
. Life
0870-6581
0870-6581
Demersalfish
fishspecies
speciesfrom
fromthe
theAzores
Azoreshave
havebeen
beenstudied
studiedsince
since1982
1982with
withthe
theaim
aimofofstock
stock
Demersal
assessment.This
Thiswork
workpresents
presentsthe
theresults
resultsfrom
froma astudy
studyconcerning
concerningthe
thegrowth
growthofofPagrus
Pagrus
assessment.
throughotolíth
otolithobservation
0 b s e ~ a t i oand
nandbybylength
lengthfrequency
frequencyanalysis.
analysis.For
Forthe
theestimation
estimation
pagrus,
pagrus
, through
theparameters
parametersofofthe
thevon
vonBertalanffy
Bertalanffygrowth
growthfunction
function,
length,three
threemethods
methodswere
were
ofofthe
, ininlength,
used, two
two direct
direct and
and one
one indirect.
indirect. The
The otolith
otolith rings
rings were
were directly
directly counted
counted byby two
two
used,
procedures: first
first byby reading
reading the
the whole
whole structure,
structure, and
and secondly
secondlybyby observing
observing a a thin
thin
procedures:
transversesection
sectionofofthe
theotolith.
otolith.Results
Resultsfrom
fromthe
thetwo
twoprocedures
procedures coincided
coincidedwith
witheach
each
transverse
other.The
Thetwo
twodirect
directmethods
methodsused
usedfor
forestimation
estimationofofthe
thegrowth
growthparameters
parametersfrom
fromthe
thehard
hard
other.
structures (otoliths):
(otoliths): direct
direct calculation
calculation and
and backcalculation
backcalculation gave
gave similar
similar results,
results,
structures
L~=103.0(1-e-0061(tf'.7)
and L~=104.6(1-e".066't+0.7))~
respectively. The
The growth
growth curve,
curve,
t=103.0(1-e· 006I (t+17)) ) and
t=104.6(1_e· o066 (t+0 7», respectively.
~=102.7(1-~-0067(t+l.0)),
similar toto the
the previous
previous ones,
ones, was
was obtained
obtained inin the
the method
method ofof
o.067(t+'O), similar
L=102.7(l_e·
indirect estimation
estimation ofof growth
growth parameters
parameters byby length
length frequency
frequency analysi
analysis
through
indirect
s through
MULTIFANsoftware.
software.The
Theasymptotic
asymptoticlengths
lengthsobtained
obtainedininthe
theprevious
previousequations
equationswere
were
MULTIFAN
cm),and
andwith
with
verylarge
largewhen
whencompared
comparedwith
withthe
themaximum
maximumobserved
observedlength
length(L(Lma0,,
7171cm),
very
m 3J< ob, ==
similarstudies.
studies.This
Thismay
mayhave
haveresulted
resultedfrom
fromincomplete
incompletesampling
samplingofofthe
theolder
olderage
ageclasses.
classes.
similar
Theresults
resultsshow
showthat
thatthis
thisspecies
specieshas
hasa aslower
slowergrowth
growthand
anda ashorter
shorterlife
lifespan
spanininAzorean
Azorean
The
watersthan
thanlhat
thatoff
offthe
theeast
eastcoast
coastofofNorth
NorthAmerica.
America.
waters
SERAFIM,MARIA
MARIAPAULA
HELENAM.
M.KRUG
KRUG1995.
1995.Contribuição
Contribuigiiopara
paraooestudo
estudo
SERAFIM,
PAULA P.P.&&HELENA
Pagruspagrus
pagrus(Linnaeus,
(Linnaeus,1758)
1758)(Pisces:
(Pisces:Sparidae),
Sparidae),nas
nas
crescimentodo
d opargo,
pargo,Pagrus
ded ecrescimento
Arquipe'lago.Ciências
CiEnciasBiológicas
Biol6gicase eMarinhas
Marinhas13A:
13A:ll-20.
Angrado
do
dosAçores.
Agores.Arquipélago.
dos
11-20. Angra
Heroismo.ISSN
ISSN0870-6581.
0870-6581.
Heroísmo.
Asespécies
esptciesdemersais
demersaisdos
dosAçores
A ~ o r etêm
st&msido
sidoobjecto
object0dedediversos
diversosestudos
estudosdesde
desde1982
1982com
corn
As
vistaà Bavaliação
avalia~Lo
desterecu
recurso.
Estetrabalho
trabalhoconsistiu
consistiunonoestudo
estudododocrescimento
crescimentodedeuma
uma
deste
rso. Este
vista
Pagruspagrus,
pagrus, através
atravCsdadaobservação
obsewa$io dos
dosotólitos
ot6litose epela
pelaanálise
andlisedede
destasespécies,
esptcies,o oPagrus
destas
frequsnciasdedecomprimento.
comprimento.Para
Paraa aestimação
[email protected] dos
dosparâmetros
pardmetrosdadacurva
curvadedecrescimento
crescimentodede
frequências
vonBertalanffy,
Bertalanffy, em
emcomprimento,
comprimento, foram
foramutilizados
utilizados três
tr&smétodos,
mttodos,dois
doisdirectos
directose e um
um
von
indirecto.AAcontagem
contagemdos
dosanéis
anCisdos
dosotólitos
otblitosefectuou-se
efectuou-sepor
pordois
doisprocessos
processosdiferentes:
diferentes:
indirecto.
observa@odos
dosotólitos
ot6litosinteiros
inteirose eobservação
obsewa@odedesecções
sec~Bes
transversaisfinas.
finas.Os
0 sresultados
resultados
observação
transversais
ambosososprocedimentos
procedimentosforam
foramcoincidentes
coincidentesentre
entresi.si.Os
0 sdois
doismétodos
mttodosdirectos
directosusados
usados
dedeambos
paraestimação
estima~go
dosparâmetros
pardmetrosdedecrescimento
crescimentoa apartir
partirdas
dasestruturas
estruturasduras
duras(otólitos):
(otblitos):
para
dos
cálculo
cdlculodirecto
direct0e erectrócalculo,
rectr6calcul0,deram
deramresultados
resultadossemelhantes,
semelhantes,LL=103.0(1-e~0."""1.7)
t=103.0(I_e· O.06I (t+I7),1.e e
o066 (t+O.7»,),respectivamente.
h=104.6(1 -e-0.066('+0.7)
respectivamente.AAcurva
curvadedecrescimento,
crescimento,L=102.7(1_e·
~ = 1 0 2 . 7 ( o067
1 - e(t+J.O»,
- ~ ~ ~1,( ~ + ' . ~ )
L=104.6(l_e·
semelhanteàs21santeriores,
anteriores,foifoiobtida
obtidapelo
pelométodo
mttododedeestimação
estima@oindirecta
indirectados
dosparâmetros
parlmetrosdede
semelhante
crescimento pela
pela análise
andlise das
das frequências
frequgncias dede comprimento
comprimento através
atravCs dodo programa
programa
crescimento
MULTIFAN.Em
Emtodas
todasasasfunções
funqBesobtidas,
obtidas,o ovalor
valordedecrescimento
crescimentoassimptótico
assimptbticofoifoimuito
muito
MULTIFAN.
cm),e e
elevadoquando
quandocomparado
comparadocom
corno ocomprimento
comprimentomáximo
miximoobservado
obsewado(Lmax
(L,, obsO ~=S=7171cm),
elevado
cornoutros
outrosestudos
estudossemelhantes.
semelhantes.Este
Estefacto
factopoderá
poderdter
terresultado
resultadodedeuma
umaamostragem
amostragem
com
11
incompleta nas
nas classes
classes etárias
etArias mais
mais velhas.
velhas. Pelos
Pelos resultados
resultados obtidos
obtidos pode-se
pode-se afirmar
afirmar que
que
incompleta
esta espécie
esptcie apresenta
apresenta um
urn crescimento
crescimento mais
mais lento,
lento, ee um
urn ciclo
ciclo de
de vida
vida mais
mais curto.
curto,
esta
comparativamente ao
ao verificado
verificado no
no costa
costa leste
lesteda
da América
Amtrica do
do Norte.
Norte.
comparativamente
Maria Paula
Paula P.
P. Serafim,
Serafim, Unidade
Unidade de
de Ciências
CiCncias ee Tecnologias
Tecnologias dos
dos Recursos
Recursos Aquáticos,
Aqudticos,
Maria
Universidade do
do Algarve,
Algarve, PT-8000
PT-8000 Faro,
Faro, Portugal.
Portugal. -- Helena
Helena Maria
Maria Noronha
Noronha Krug,
Krug,
Universidade
Departamento de
de Oceanografia
Oceanografa ee Pescas,
Pescas, Universidade
Universidade dos
dos Açores,
Apres, PT-9900
PT-9900 Horta,
Horta,
Departamento
Portugal.
Portugal.
INTRODUCTION
INTRODUCTION
Pagrus pagrus (Linnaeus,
(Linnaeus, 1758)
1758)
The red
red porgy
porgy Pagrus
The
in Azorean
Azorean waters
waters isis aa bottom
bottom dwelling
dwelling species
species
in
common from
from temperate
temperate to
to subtropical
subtropical habitats.
habitats.
common
This demersal
demersal species
species has
has been
been studied
studied in
in the
the
This
Department of
of Oceanography
Oceanography and
and f'isheries
Fisheries since
since
Department
1982, because
because of
of their
their high
high economic
economic value
value and
and
1982,
also for
for their
their vulnerability
vulnerability to
to fishing.
fishing. The
The Azores
Azores
also
do not
not have
have aa continental
continental shelf,
shelf, and
and its
its shallow
shallow
do
water isis Iimited
limited to
to some
some fishing
fishing banks
banks and
and sea
sea
water
mounts. The
The Azorean
Azorean demersal
demersal community
community isis
mounts.
very diverse
diverse but
but the
the intensive
intensive fishing
fishing has
has lead
lead to
to
very
decrease in
in stock,
stock, in
in this
this way
way endangering
endangering the
the
aa decrease
(SILVA
et aI.
al. 1994).
1994).
fisheries viability
viability (SILVA
et
fisheries
The red
red porgy
porgy isis an
an economically
economically important
important
The
species for
for the
the fisheries
fisheries in
in the
the Azorean
Azorean
species
Archipelago. The
The demersal
demersal fisheries
fisheries include
include 20
20
Archipelago.
Pagrus
species and
and there
there isis no
no specific
specific fishery
fishery for
for Pagrus
species
pagrus. It
It isis found
found at
at depths
depths ranging
ranging from
from 50
.% toto
pagms.
150 m
m and
and isis captured
captured with
with long-Iine.
long-line.
150
It isis important
important to
to study
study these
these species
species in
in order
order
It
to implement
implement management
management policies
policies and
and rational
rational
to
exploitation of
of the
the stock.
stock. Th
The
overall objective
objective of
of
e overall
exploitation
this study
study was
was to
to obtain
obtain basic
basic information
information to
to
this
fisheries management
management in
in the
the Azorean
Azorean waters.
waters.
fisheries
The specific
specific objective
objective was
was to
to investigate
investigate if
if the
the
The
red porgy
porgy can
can be
be aged
aged by
by lhe
the number
number of
of
red
deposition rings
rings on
on otoliths
otoliths and
and apply
apply different
different
deposition
methods of
of growth
growth analysis
analysis such
such as
as direct
direct
methods
reading and
and backcalculation,
backcalculation, and
and indirect
indirect
reading
methods Iike
like length-frequency
length-frequency analysis
analysis using
using the
the
methods
MULTIFAN
software.
MUL
TIF AN software.
MATERIAL
AND METHODS
METHODS
MA
TERIAL AND
The data
data were
were obtained
obtained from
from a
The
Department
of
Oceanography
Oceanography
Department
12
project at
at the
and
Fisheries
and Fisheries
(DOF), University
University of
of the
the Azores,
Azores, Horta,
Horta, for
for the
the
(DOF),
studyof
of demersal
demersal species
species in
in Azorean
Azorean waters.
waters. The
The
study
fish samples
samples were
were obtained
obtained from
from monthly
monthly
fish
sampling
of
the
commercial
longline
captures
sampling of the commercial longline captures
and
from
the
catches
by
the
RN
"Arquipdago
and from the catches by the RJV "Arquipélago".".
Biological samplling
sampling
Biological
In ali
all specimens
specimens the
the fork
fork length
length (FL)
(FL) was
was
ln
measured
to
the
nearest
cm
below,
and
sexes
measured to the nearest cm below, and sexes
macroscopically determined
determined (functional
(functional males,
males,
macroscopically
functional
females,
transitionals
and
immature
functional females, transitionals and immature
juveniles). Red
Red porgy
porgy sampled
sampled ranged
ranged from
from 15
15 to
to
juveniles).
63
cm
FL
(Table
I),
which
corresponds
to
the
63 cm FL (Table 1), which corresponds to the
size range
range of
of red
red porgy
porgy caught
caught in
in the
the longline
longline
size
fisheries.
fisheries.
Otolith co!!ection
collection and
and processing
processing
Otonith.
Sagittae were
were collected
collected during
during the
the period
period 199
1991
to
Sagittae
1 to
1993. Altogether,
Altogether, 378
378 were
were at
at our
our disposal,
disposal, of
of
1993.
which 358
358 were
were easy
easy to
to read.
read. The
The sagittae
sagittae were
were
which
removed, c1eaned
cleaned and
and dried.
dried. For
For age
age reading
reading they
they
removed,
were immersed
immersed in
in alcohol
alcohol and
and observed
observed using
using aa
were
binocular microscope
microscope (magnification
(magnification 16x)
16x) with
with
binocular
transmitted light
light and
and an
an image
image analysis
analysis system
system
transmitted
(NHI-ImageO, v.l.44,
v. 1.44, on
on Macintosh®
Macintosh@ Quadra
Quadra
(NHI-Image©,
800).
800).
The otoliths'radii
otoliths'radii (OR)
(OR) were
were measured
measured using
using the
the
The
longest
otolith
axis.
The
corresponding
radius
longest otolith axis. The corresponding radius of
each annulus
annulus is
is the
the distance
distance between
between the
the nucleus
nucleus
each
and the
the beginning
beginning of
of each
each opaque
opaque zone.
zone. The
The
and
larger otoliths
otoliths were
were processed
processed in
in the
the laboratory
laboratory
larger
by making
making aa thin
thin transverse
transverse section
section placed
placed in
in
by
polyester resin
resin according
according to
to the
the method
method described
described
polyester
by BEDFORD
BEDFORD(1983).
(1983). After
After testing
testing the
the annual
annual
by
periodicity of otolith
otolith rings by analysis
analysis of
periodicity
marginal increments,
increments, the
the number
number of rings could
could
marginal
Table 1
Fork length (FL) distributions of the age groups of P. pagrus and mean values at each age class and
standard deviations (SD).
Age class (years)
then be converted into ages. For Pagrus pagrus,
when these observations were made, the first
translucent zone was interpreted .as representing
the passage of the fish through the second winter.
These pooled samples were used to calculate
the von Bertalanffy growth parameters (k and
Lw) through the MULTIFANO software (OTTER
RESEARCH
1992), that is being commonly applied
to fisheries data (e.g. FOURNDER
et al. 1990).
Analysis of otoliths data
To start MULTIFAN, the results of otolith
readings, as well as information from literature,
The application of the direct method was based were used. For the initial searches (unstructured
on sagittae and fork length (FL) data, under the samples) the following options were introduced:
assumption that ring counts indicate total age (in range of k values from 0.055 to 0.080 (with
years). The von Bertalanffy parameters obtained increments of 0.005); range of annual classes
were estimated using the non-linear least squares from 11 to 17 years; limits for the first cohort, 15
fitting method of FISHPARM program ( S ~ etA cm (lower) and 18.7 cm (upper) with a standard
al. 1988). The van Bertalanffygrowth parameters deviation of 0.97 cm. The subsequent modified
were also estimated by backcalculation using the systematic searches were performed in order to
FL-OR (fork length /otolith radius) relation of actively find the error of the first length
the Fraser-Lee method (FRANCIS1990) (for the (introduced by fishing gear selectivity), as well as
calculation of the age-length key). This to find the standard deviations for the mean
relationship W O R ) was calculated from 355 length for each age class. For some fish
specimens by geometric regression. The age- populations, variations in length-at-age is not
length key, was done by direct reading of the constant across cohorts. This hypothesis allows
otoliths and the mean lengths of each age group the standard deviation of length-at-age to
were estimated. The von Bertalanffy growth increase or decrease linearly with age. New
parameters were then estimated through the non- searches of k values, outside the initial range,
linear least squares fitting method of the were also done after the previous operations.
FISHPARM program ( S ~ etAa]. 1988).
To obtain the best k and L,parameters, the
Collection and analysis of length-frequency values of the maximum-likelihood function of
each systematic search obtained, were compared
data
Length data (measurements done as mentioned trough x2 test.
Finally, after obtaining the best fitted k and
before) of the most important commercial fish
L,,
values, the toparameter was calculated by the
species of the Azores, collected in several
(GULLAND1971):
equation
harbours, is available in the Regional Fish
Sampling Database Program of DOF. Length
data of unsexed red porgy of this database, for
the period from January 1987 to December 1993,
where Lt is the length on time t.
was selected for this study (totaI of 3351
individuals). However, a part of these data could
not be used directly in the analysis as several
monthly samples were composed by only a few RESULTS
specimens. To avoid this problem of insufficient
sample size, the data of fish captured during the Period of annulus formation
late autumn and winter months (November to
April), when growth is slower, were pooled The periodicity in otolith zonation, through
together. Four samples of pooled six-month observation of their optical nature with respect to
length-frequency data, each one with more than the marginal area and by mean monthly marginal
300 specimens, were established by this increments, agrees with the hypothesis of annual
formation (Fig. 1).
procedure.
I
.
Jan Feb Mat A ~ IMay h n Jul Aug Set Od Nov Deo
Months
Fig. 1. Mean monthly marginal increments (%).
The covariance analysis (ANCOVA)
(SOKAL& ROHLF1981) detected no
significant differences between sexes
in
growth
parameters.
The
backcalculated age-length key is
presented in Table 3. From these, the
estimated growth parameters of von
Bertalanffy equation are presented in
Table 4.
The von Bertalanffy parameters were
estimated for each sex (males and
females) (Fig. 2).
S.D.-standard G~~~~ analysis based on length-
deviation; Bias = standard error.
frequencies
The mean marginal increments were highest
in February, November and December. In March
to May the mean marginal increments were very
low. Thus, the opaque edges were formed from
early March and late May in each year (Fig. 1).
This agree with an annual periodicity of the
growth zone formation.
The best options of the model structure chosen
for the red porgy length frequency data set
included age dependent standard deviation in
length at age and first length bias. The analysis
included 50 length classes. Parameter estimates
of the means and standard deviations of length at
age are given in Table 5.
Table 2
Growth in length
Growth analysis bassd on otoliths-ring count
Growth parameters, standard deviation (SD) and
coefficient of variation (CV), based upon direct
reading of otoliths.
The von Bertalanffy model was fitted to the
aggregate otoliths-ring count data for relative age
Parameters estimates
SD
classes 2-1 1. The data for the relative age class 1
L, 103.0cm
0.186
were excluded because they corresponded to fish
k
0.061
0.018
to
-1.7year
0.458
with more than one year
of age. This is associated
70
with
bias due to
%
'
(Females) LF=92.6 ( ~ - e ~ . ~ ' ~ I' + ' - ~
0
selectivity of the gear.
60 \ (Males) LF=lOI.I ( ~ - e " ~ ~ ~ ~ (3' + ' ' "
The data for relative age
classes 12 and 13 were
50
excluded due to their
40
small sample size. From
30
these data the von
Bertalanffy
growth %
parameters were obtained
20
(Table 2).
10
The predominance of
o
females was observed in
0
1
2
3
4
5
6
7
8
9
1 0 1 1
younger Iength classes
while males reached
Age classes (years)
greater length classes. Fig. 2. The Von Bertalanffy curve fitted to males and females.
CV
0.18%
29.5%
-26.9%
0
0
5
2
B
1 2 1 3 1 4
Table 3
Mean length obtained by backcalculation, mean values at each age class and standard deviations (SD).
1
2
3
4
5
6
7
8
9
10
11
12
SD
10.95
10.73
11.23
11.31
10.96
10.93
11.00
11.62
11.34
11.19
11.64
11.36
10.80
11.16
0.29
16.92
17.78
17.63
17.18
17.47
17.30
17.82
18.29
18.59
18.23
19.40
17.81
17.87
0.68
23.64
23.09
22.59
22.76
22.53
23.41
23.44
23.30
23.92
25.34
23.62
23.42
0.78
28.55
27.98
27.86
27.38
28.20
28.68
28.74
29.36
30.15
29.13
28.60
0.80
33.43
32.95
32.25
32.60
33.10
33.45
34.56
34.79
33.46
33.40
0.83
37.59
36.73
36.97
37.00
37.71
38.87
38.92
38.32
37.76
0.86
41.12
40.63
41.07
42.15
42.44
42.00
43.36
41.82
0.95
44.61
44.60
44.85
46.47
45.64
46.77
45.49
0.96
47.88
48.27
50.14
48.90
50.50
49.14
1.14
51.33
53.38
53.38
54.31
53.10
1.26
56.20
55.91
58.10
56.74
1.19
59.86
59.81
59.83
0.04
N
755
776
121
221
117
63
45
17
17
7n
~n
A
Number of rings
Age class
1
2
3
4
5
6
7
8
9
10
11
12
13
FLmean(cm)
Table 4
Table 5
Growth parameters, standard deviation (SD),
coefficient of variation (CV), based on mean
backcalculated lengths.
Parameter estimates SD
CV
0.050
0.05%
L, 104.6 cm
7.6%
k 0.066
0.005
to -0.7 year
0.09
-12.9%
Von Bertalanffj parameters, k, L, & b, mean fork
lengths (FL), standard deviations (SD) at each age class
in the period considered.
I+
K(year-9
to
N
?
In the period analysed (1987-1993), the best
fit corresponded to the age classes 12 and 14 for k
values 0.067 and 0.072, respectively. For a k
value of 0.067 (age class 12) the model, when
four parameters are added, gives a value of 27.35,
which represents the double of what was obtained
for the model of age class of 11 years. On the
other hand and with the same number of
parameters added, there are an increase of 43.07
in the maximum-likelihood function, when the age
class of 14 years (k=0.072) is considered in
relation to the model age class of 13 years.
The ~2 test, with four degrees of freedom and
probability level 0.90, gives a value of 7.78,
which means the both models should be accepted
as significant, neither being rejected. However,
the model with 12 age classes was selected as the
best fit due to its low value of the maximum-
102.7
Age class
(years)
1
2
3
4
5
6
7
8
9
10
11
. 12
13
0.067
Mean FL
(cm)
-0.96
2126
0.99
SD
(cm)
-------
------
18.52
24.00
29.11
33.90
38.37
42.56
46.47
50.13
53.54
56.74
59.73
62.53
1.88
1.73
1.61
1SO
1.40
1.31
1.24
1.17
1.12
1.06
1.02
0.98
likelihood function.
The predicted aggregate length frequency
distribution fitted the observed distribution very
well over the entire range of size (Fig. 3) and the
predicted mode closely matched the actual modes
NOV-APR
1987-1988
N= 341
NOV-APR
1989-1990
NOV-APR
1990-1991
NOV-APR
1991-1992
Fig. 3. The result of the best fit of MULTIFAN, for P. pagrus length-frequency data set, where the
vertical lines represent the estimated mean length of these age classes.
in the four groups of samples pooled by six
months intervals, which correspond to the lowest
period of 'pecies growth (Nov-Apr). The
predicted modal distribution pattern indicates
that there were usually 12 predominant age
classes in the longline catch sample.
Comparison of growth curves
W, = 0 . 2 2 . ~ ~ O . (&0.98)
~~
~
~the L- in
~ the length
l hnction
~ by
W-, the following equation was derived for the
growth in weight using the direct method (otolith
reading).
W, = 26.72(1- e
The comparison of the von Bertalanffy growth DISCUSSION
curves estimated from length-frequencies, by
direct reading of otoliths andby backcalculatio~,
From this study one can assume that the otolith
were similar (Fig. 4).
(sagina) age reading is a
valid method due to the
alternate formation of
opaque and translucent
bands in each year. The
results of the three
different methods of
analysis were similar.
Growth parameters were
estimated by MANOOCH
& HUNTSMAN(1977)
and ROUMILLAT &
WALTZ (1993) from
01
catches, off the North
3
4
5
6
7
8
9 1 0 1 1 1 2 1 3 1 4
0
1
2
Carolina coast and the
Age classes (years)
south-eastern coast of
fig. 4. The Von Bertalanffy curves fitted by three different methods..
North
America,
In spite of substantial differences in the L,
respectively (Fig. 5). MANOOCH& HUNTSMAN
and k estimates, covariance analysis (ANCOVA) (1977), considered that k was overestimated and
(SOW & ROHLF 1981) was applied and then that L
C
- was underestimated because they were
compared to the slope of three growth functions. estimated by backcalculation using the number of
These
results indicate that growth curve rings on scales (k=0.096; Lm=76.3 cm; to=-1.88)
estimates are significantly similar for F (2,31) and they found one 74 cm individual.
with p<0.05. The utilisati~n of the Newrnan- R O ~ J , T & WALTZ (1993) based his
allows the calculation on otolith reading but did not present
Keuls test @ O W & R O ~ ~ F
pairwise cornparation of ~ - ~ ~ FL
e a nfor the
growth parameters, only mean lengths at
methods, showing diffeTable 6
rence in MULTIFAN for
p<0.05 (Table 6).
Covariance test (Newman-Keuls) for the analysed methods
back-calculation direct reading MULTIFAN
Growth in weight
40.03
39.80
42.98
---0.860
0.025"
From the length(cm)- back-calculation
weight(g) relation, the direct reading
0.860
---0.590
following function was MULTIFAN
0.025"
0.590
---estimated:
*- significant a p<0.05.
~
i
otolith reading in the age
class of more than 7
years.
/:'
60
../$
*
Despite this concluMANNOCH (1977)
9
* ROUMILLLAT (1993)
sion, the results should
E
be considered with some
5 40
U)
caution, mainly because
*
C
../
s30
of undersampling in the
Y,
./
younger (0-1 age class)
2 20
../
and older age classes
/"
10
(over 9 years). An
interesting study in this
0L
0
1
2
3
4
5
6
7
8
9 1 0 1 1 1 2 1 3 14
line of work would be the
Age classes (years)
growth
of
young
specimens in captivity,
Fig. 5. Growth curves obtained compared with other authors.
since their capture in the
different age classes, as shown in figure 5. The fishery is limited, caused by the selectivity of the
largest individual was 55.7 cm in total length
gear.
70
(analysis based on length-frequencies)
-:.(backcalculation)
- (direct method)
-
"
(TL).
The predominance of females in younger
length classes while males reach greater length
ACKNOWLEDGEMENTS
classes is in agreement with the sexual
characteristic of the species (MANOOCH1976; We would like to thank the support given by
& HUNTSMAN
1977; ROUMILLART
& Department of Oceanography and Fisheries,
MANOOCH
University of Azores. Special thanks is due to Dr
WARTZ1993; VAUGHAN
et al. 1992).
The growth of red porgy in the Azores is Malcolm R. Clarke and Dr. Helen R. Martins for their
critical review of the manuscript and for revision of
slower than their equivalents of the American
the English language.
Atlantic coast. For the asymptotic length (Lw)
the population of the Azores has generally larger
values than those found on the Atlantic coast of REFERENCES
North America, but this parameter can be
influenced by the number of observed age classes BEDFORD,
B.C. 1983. A method for preparing sections
and by the sampling of older age classes, since
of large numbers of otoliths embedded in black
the absence of older individuals can cause the
polyester resin. Journal du Conseil Internationale
overestimation of this parameter.
pour 1'Exploration de la Mer 41:4-12.
D.A., J.R. SIBERT,J. MNKOWSKI
& J.
Although it was not possible to determine the FOURNER,
HAMPTON
1990. MULTIFAN a likelihood-based
age boundaries in the Azorean population, it is
method for estimating growth parameters and age
clear that this species has a slow growth and
composition for multiple length frequency data
short lifespan. In this analysis 13 age classes
sets illustrated using data for southern bluefin tuna
were observed while MANOOCH& HUNTSMAN
(Thunnus maccoyii). Canadian Journal of
(1977) observed 15 age classes in U.S. waters.
Fisheries and Aquatic Sciences 47: 301-3 17.
The largest specimen captured from this study FRANCIS,
R.I.C.C. 1990. Back-calculation of fish
was 71 cm FL, a big difference when compared
length: a critical review. Journal of Fish Biology
to the maximum length Lw estimates. This fact
36:883-902.
can be explained by the bias in subsampling GULLAND,
J.A. 1971. Manual de Mktodos para la
Evaluacio'n de las Poblaciones de Peces. FAO,
older age classes, which can lead to an
Roma. Editorial Acribia, Zaragoza, Espaiia. 164
overestimation of the Lw. Another important fact
PP.
is related to a possible overestimation of the
MANOOCH, C.S. III 1976. Reproductive cycle,
fecundity and sex ratio of red porgy, Pagrus
pagrus (Pisces: Sparidae) in North Carolina.
Fishery Bulletin 74(4):775-781. .
MANOOCH,C.S. TII & G.R. HUNTSMAN1977. Age,
growth and mortality of the red porgy, Pagrus
pagrus. Transactions of the American Fisheries
Society 1O6(l): 26-33.
OTTERRESEARCH.1992. MULTIFAN 3: User's Guide
and Reference Manual. 65 pp.
ROUMILLAT,
W.A. & C.W. WALTZ1993 Biology of the
Red Porgy, Pagrus pagrus, from the Southeastern
United States. Unpublished Data Report 1993,
MARMAP, South Carolina Wildlife and Marine
Resources Departement. 26 pp.
SAM, S.B., C.W. RECKSIEK
& M.H. PRAGER1988.
Fishery Science Application System. A compen-
dium of microcomputer programs and manual of
operations. American Elsevier. 223pp.
SILVA,H.M., H.M. KRUG,& G.M. MENEZES,1994.
Bases para a Regulamentag20 de Pesca de
Demersais nos Aqores. Arquivos do DOP, S&ie
Estudos, No 4. Universidade dos A~ores,Horta. 41
PP.
SOKAL,R.R. & F.J. ROHW 1981. Biometry. W.H.
Freeman & Co.. San Francisco. 859 pp.
III.
VAUGHAN,
D.S., G.H. HUNTSMAN,
C.S. MANOOCH
F.C. ROHDE & G.F. ULRICH 1992. Population
characteristics of the red porgy, Pagrus pagrus,
stock off the Carolinas. Bulletin of Marine Science
50(1): 1-20.
Accepted 29 November 1995.