MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
https://doi.org/10.47193/mafis.3712024010103
ABSTRACT. The Argentine anchovy, Engraulis anchoita, plays a vital role as a key prey
species for several marine predators in the north Patagonian marine ecosystem of the Atlantic
Ocean. Reconstructing the length and weight of each consumed specimen is essential to provide a
detailed description of the trophic ecology of top marine predators. Predictive linear regression
equations were calculated for the Patagonian stock of Argentine anchovy to estimate parameters of
length-weight relationships using measurements of whole individuals and diagnostic elements such
as otoliths, head bones and pectoral fin bones. Among the diagnostic elements analyzed, the clei-
thrum and dentary exhibited the best fit. This study validates the use of head and pectoral girdle
bones as reliable indicators for predicting the weight and length of Argentine anchovy across a wide
size range, which corresponds to the target range of various predators. These relationships can con-
tribute to the determination of body condition, estimation of consumed biomass, and calculation of
energy density, providing valuable insights into the trophic ecology of predators in the southern
Atlantic Ocean.
Key words: Head bones, otoliths, regression, Patagonia, measurements, length-weight.
Utilización de los huesos de la cabeza, cintura pectoral y otolitos para estimar la talla y el peso
de la anchoíta (Engraulis anchoita), especie clave en el ecosistema marino patagónico
RESUMEN. La anchoíta, Engraulis anchoita, juega un papel vital como especie presa clave para
varios depredadores marinos en el ecosistema marino patagónico norte del Océano Atlántico.
Reconstruir la longitud y el peso de cada espécimen consumido es esencial para proporcionar una
descripción detallada de la ecología trófica de los principales depredadores marinos. Se calcularon
las ecuaciones de regresión lineal predictiva para el stock patagónico de anchoíta, a fin de estimar
los parámetros de las relaciones talla-peso utilizando medidas de individuos completos y elementos
de diagnóstico como otolitos, huesos de la cabeza y huesos de la aleta pectoral. Entre los elementos
diagnósticos analizados, el cleitro y el dentario exhibieron el mejor ajuste. Este estudio valida el uso
de los huesos de la cabeza y la cintura pectoral como indicadores confiables para predecir el peso y
la talla de la anchoíta en un amplio rango de tamaños, que corresponden al rango objetivo de varios
depredadores. Estas relaciones pueden contribuir a la determinación de la condición corporal, la
241
*Correspondence:
sfernandez@conicet-cenpat.gob.ar
Received: 5 June 2023
Accepted: 26 July 2023
ISSN 2683-7595 (print)
ISSN 2683-7951 (online)
https://ojs.inidep.edu.ar
Journal of the Instituto Nacional de
Investigación y Desarrollo Pesquero
(INIDEP)
This work is licensed under a Creative
Commons Attribution-
NonCommercial-ShareAlike 4.0
International License
Marine and
Fishery Sciences
MAFIS
NOTE
Using head, pectoral girdle bones and otoliths to estimate length
and weight of Argentine anchovy (Engraulis anchoita), a key species in
Patagonian marine ecosystem
SANTIAGO J. FERNÁNDEZ1, 2, *, CYNTHIA IBARRA1, XIMENA NAVOA3and JAVIER E. CIANCIO1
1Centro para el Estudio de Sistemas Marinos (CESIMAR), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Centro
Nacional Patagónico (CENPAT), Boulevard Almirante Brown 2915, U9120ACD - Puerto Madryn, Argentina. 2Facultad de Ciencias Naturales
y Ciencias de la Salud, Universidad Nacional de la Patagonia San Juan Bosco, Boulevard Almirante Brown 3051, U9120ACE - Puerto
Madryn, Argentina. 3Instituto de Investigación de Hidrobiología, Facultad de Ciencias Naturales y Ciencias de la Salud, Universidad Nacional
de la Patagonia San Juan Bosco, Gales 48, U9100CKA - Trelew, Argentina.
ORCID Fernández J. Santiago https://orcid.org/0000-0002-2805-0865, Cynthia Ibarra https://orcid.org/0000-0001-9561-5531,
Javier E. Ciancio https://orcid.org/0000-0003-1674-842X
Diet studies are crucial for understanding
trophic dynamics, foraging behaviors and life his-
tory patterns of most species within ecosystems
(Hódar 1997). Descriptions in terms of prey size
are essential for determining predator consump-
tion rates, total prey biomass consumed, selectiv-
ity of a predator towards a specific prey size, and
modeling consumer energy use (Hansel et al.
1988). Additionally, the diet of marine mammals
and seabirds can provide early indications of fluc-
tuations in fish populations, making them sentinel
species for marine ecosystem variability (Velarde
et al. 2013; Ciancio et al. 2021; Pirotta et al.
2022; Ramos and Furness 2022).
Direct measurement and weighing of prey
items is challenging due to fragmentation or
digestion (Hódar 1997). Therefore, reconstruct-
ing the length or weight fragmented preys in diet
samples is often necessary. Despite advanced
digestion in predators, certain prey parts exhibit
slower digestion and maintain a constant relation-
ship with prey body size, enabling reliable identi-
fication and reconstruction of most prey items
(Hansel et al. 1988). This is the case of bones, for
which several studies use marine fish diagnostic
bones to estimate original lengths and weights of
prey consumed by predators through regressions
(Sinovčić et al. 2004; Tapella and Lovrich 2006;
González-Zevallos et al. 2010; Pérez Comesaña
et al. 2014; Riestra et al. 2020).
In the marine ecosystem of northern Patagonia,
the dominant schooling fish is the Patagonian
stock of Argentine anchovy, Engraulis anchoita
(Hubbs and Marini, 1935). This stock is distributed
from 41° S to 48° S, and a large fraction is located
in front of Península Valdés during the last quarter
of the year in waters ranging from 50 to 80 m deep
(Hansen 2004). They play a key role as an impor-
tant prey item for several fishes, seabirds, and
marine mammals (Crespo et al. 1997; Koen Alon-
so et al. 2000; Gatto and Yorio 2009; Belleggia et
al. 2012; Loizaga de Castro et al. 2016; Ibarra et al.
2018; Fernández et al. 2019). To provide a detailed
description of the trophic ecology of top predators,
it is essential to reconstruct the length and biomass
of each consumed specimen. Therefore, our goals
were to estimate parameters of length-weight rela-
tionships using measurements of whole individu-
als, as well as diagnostics elements such as
otoliths, head bones and pectoral fin bones from
the Patagonian stock of Argentine anchovy.
A total of 125 specimens from the Patagonian
stock of Argentine anchovy were studied. Sixteen
specimens were captured in April 2013 using a
net in Golfo Nuevo, Península Valdés, Chubut
Province. Additionally, specimens from commer-
cial fishery vessels in Rawson, Chubut Province
during October 2014 (n =23), December 2019 (n
=28), and June-July 2021 (n =41) were included.
All specimens were frozen for further analysis.
Finally, 17 samples preserved in formalin from
the same stock were obtained from the Ichthy-
ological Collection of the Centro Nacional
Patagónico (CENPAT) (n =17, CNPICT 1992/75,
CNPICT 1992/76 and CNPICT 2000/23).
Furca length (FL) from individuals was meas-
ured to the closest millimeter with an ichthyome-
ter and weighted (W) in grams using an analytical
balance. For samples obtained since 2019, total
length (TL) was also measured (n =69). Furca
length refers to the measurement from the tip of
the longest jaw of a fish to the center of the fork
in the caudal fin, while total length is measured to
the longest caudal lobe.
Dissection techniques were employed to disar-
ticulate anchovies after they have been soaked in
boiling water (100 °C) for at least ten minutes.
Selected diagnostic elements (Table 1) were those
242 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
estimación de la biomasa consumida y el cálculo de la densidad de energía, proporcionando información valiosa sobre la ecología trófica
de los depredadores en el Océano Atlántico sur.
Palabras clave: Huesos de la cabeza, otolitos, regresión, Patagonia, mediciones, talla-peso.
that commonly appear in stomach contents based
on Gosztonyi and Kuba (1996) and our personal
experience. Photographs of elements were taken
using a digital camera with a scale. Images were
subsequently digitally measured with ImageJ
1.48v software (Schindelin et al. 2015) according
to the description in Table 1 and the scheme
delineated in Figure 1.
Statistical analyses were conducted using the R
software (R Core Team 2021). Relationships were
modeled using ‘lm’ function from car package
(Fox et al. 2007). Normality and homoscedastici-
ty assumptions were tested using graphical meth-
ods. Prior to model fitting, log-transformed data
plots were used to detect outliers following the
suggestion of Froese (2006). The performance of
the regression models was assessed using the root
mean squared error (RMSE) during repeated K-
fold cross-validation (10 folds with 100 repeats)
(Linhart and Zucchini 1986). The RMSE provides
a measure of the average difference between the
observed and predicted values of the model,
allowing for the evaluation of how well the model
fits the data. In our study, RMSE values that were
a small fraction of the data (less than 10%) were
considered as low, indicating a good fit of the
model and accurate predictions.
Length-weight relationship was examined
using the equation W= aLb(Froese 2006), where
Wrepresents the wet weight, Lis the furca
length or total length (TL, when data was avail-
able), ais the intercept, and bis the scaling
exponent. To establish the relationship, the expo-
nential model was linearized by applying loga-
rithmic transformation (log[W] =log [a] +b
log[L]). The scaling exponent (b) of the length-
weight relationship determined the growth pat-
tern, indicating isometric growth (b = 3), posi-
tive allometric growth (b >3), or negative allo-
metric growth (b <3). The growth status was
determined using a t-test according to the b=3
isometric growth hypothesis (Sokal and Rohlf
1987), providing insights into the species’
growth patterns and indicating its status in a
given environment (Froese 2006). Total length is
more widely used than furca length in fisheries
research (Carlander and Smith 1945). Therefore,
establishing a more precise relationship between
these two measurements was aimed for lengths
conversion. The equation TL =a+b L, with TL
and Lin mm, was used.
Simple linear regression equations were calcu-
lated to estimate the furca or total length versus
the length of diagnostic elements using the equa-
243
FERNÁNDEZ ET AL.: DIAGNOSTIC ELEMENTS REGRESSIONS FOR ARGENTINE ANCHOVY
Table 1. Measurement description for the diagnostic elements of the Argentine anchovy.
Measurement Abbreviation Description
Parasphenoid LPar Length from the tip of the anterior process to the posterior incisure
Dentary LDen(1) Length from the rostral tip to the external incisure
LDen(2) Maximal length measured from the rostral tip to the caudal tip of the
ventral process
Maxilla LMax Maximal length, from the rostral tip of the external process to the tip of
the caudal process
Anterior ceratohyal LCer Maximal length of the anterior ceratohyal
Cleithrum LCle Length from the dorsal to the ventral tip
Otolith LOto(1) Maximal length (rostro-cauda axis)
LOto(2) Maximal width (dorso-ventral axis)
tion L =a +b DL, where DL is the length (mm)
of the diagnostic element. Additionally, linear
regression on log-transformed data was used to
determine relationships between wet weight and
the length of diagnostic elements. An ANCOVA
was performed to test for significant differences
between left and right measurements of pair
bones and otoliths, as they may not always pro-
vide the same estimation of fish length (e.g.
Martínez‐Polanco et al. 2022).
Population studies comprised Argentine
anchovies with furca lengths ranging from 40 to
170 mm (mean ±SD: 120.50 ±30.89 mm, n =
125), total lengths between 87 and 195 mm (mean
±SD: 159.92 ±24.75 mm, n =69), and wet
weight that ranged from 0.23 to 47.54 g (mean ±
SD: 20.57 ±13.62 g, n =125) (Table 2).
Length-weight relationships for furca and total
length were statistically significant (R2=0.989
and 0.922, respectively). Estimated values of b(b
±SE: bL=3.69 ±0.03 and bTL =3.41 ±0.12) dif-
fered significantly from 3 (t-test, p-values <
0.05), indicating positive allometric growth in our
samples. The linear relationship between total
length and furca length was also significant (R2>
0.914, p-value <0.05) (Table 2).
Fifteen linear regressions were conducted to
estimate furca and total length using the length of
244 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
Figure 1. Scheme of measurements performed on cranial bones, pectoral girdle bones, and otoliths of the Argentine anchovy. A)
Fish body measurements. B) Parasphenoid and the left paired bones (from left to right): dentary, maxilla, anterior cera-
tohyal, and cleithrum. C) Left otolith indicating the two measurements used. 1) LPar: length of the parasphenoid. 2)
LDen(1): length of dentary. 3) LDen(2): maximal length of dentary. 4) LMax: maximal length of the maxilla. 5) LCer:
maximal length of the anterior ceratohyal. 6) LCle: length of cleithrum. 7) LOto(1): maximal length of otolith. 8)
LOto(2): maximal width of otolith. Measurements of right bones and otoliths were taken in the same manner as shown
in the diagram.
A
BC
Total length
Furca length
0 20 mm
0
20
mm
0
10
mm 1
2
3
4
5
6
7
8
245
FERNÁNDEZ ET AL.: DIAGNOSTIC ELEMENTS REGRESSIONS FOR ARGENTINE ANCHOVY
Table 2. Parameters of morphometric equations calculated for the Argentine anchovy. Mean, standard deviation, minimum and maximum size of the fish
(lengths in mm, weight in g), R2, and the root mean squared error (RMSE) value are presented. The level of significance is not shown since all equa-
tions had a p-value <0.05. For length equations, use the formula L= a+b* DL, and for weight equation, use W=aDLb, where Lcan be total length
(TL) or furca length (FL), and DL is the length of the diagnostic element. To estimate the relationship between total length and fork length, use the
equation TL =a+b* FL.
Diagnostic Mean SD Min. Max. N Relationship Parameters [CI 95%inf, CI 95%sup] R2 RMSE
element
a b
Furca length 120.50 30.89 40 170 125
(FL)
Total length 159.93 24.76 87 195 69 TL versus 13.787 1.074 0.914 6.966
(TL) FL [2.726, 24.858] [0.994, 1.154]
Wet weight 20.57 13.62 0.23 47.57 125 W versus 3.31 ´ 10-7 3.692 0.989 0.113
(W) FL [2.41 ´ 10-7, 4.55 ´ 10-7] [3.625, 3.759]
69 W versus 7.54 ´ 10-7 3.415 0.922 0.168
TL [2.22 ´ 10-7, 2.55 ´ 10-6] [3.173, 3.656]
Parasphenoid 13.06 3.69 5.65 20.87 68 LPar versus -13.206 10.087 0.934 6.461
TL [-24.639, -1.772] [9.425, 10.748]
122 LPar versus -12.841 8.649 0.949 6.647
FL [-18.590, -7.093] [8.287, 9.012]
LPar versus 2.51 ´ 10-4 4.038 0.947 0.241
W [1.50 ´ 10-4, 4.00 ´ 10-4] [3.865, 4.211]
Dentary left 15.41 3.5 6.46 20.44 69 LDenL(1) -13.137 10.106 0.887 8.192
(1) versus TL [-28.317, 2.044] [9.228, 10.985]
123 LDenL(1) -7.287 8.373 0.971 5.275
versusFL [-11.355, -3.219] [8.114, 8.633]
LDenL(1) 5.34 ´ 10-4 3.778 0.978 0.167
versus W [4.00 ´ 10-4, 7.00 ´ 10-4] [3.676, 3.879]
Dentary left 19.26 4.51 7.71 26.65 69 LDenL(2) -3.998 7.635 0.888 8.211
(2) vs. TL [-18.342, 10.345] [6.973, 8.296]
123 LDenL(2) -4.524 6.553 0.977 4.852
versus FL [-8.101, -0.947] [6.372, 6.735]
LDenL(2) 2.77 ´ 10-4 3.714 0.978 0.167
versus W [2.10 ´ 10-4, 3.70 ´ 10-4] [3.615, 3.814]
Dentary right 15.27 3.68 4.4 20.65 69 LDenR(1) -11.649 9.978 0.889 8.192
(1) versus TL [-26.551, 3.253] [9.119, 10.837]
125 LDenR(1) -6.017 8.286 0.973 5.094
versus FL [-9.871, -2.164] [8.041, 8.532]
LDenR(1) 5.98 ´ 10-4 3.736 0.978 0.166
versus W [4.60 ´ 10-4, 7.80 ´ 10-4] [3.637, 3.836]
246 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
Table 2. Continued.
Diagnostic Mean SD Min. Max. N Relationship Parameters [CI 95%inf, CI 95%sup] R2 RMSE
element
a b
Dentary right 19.10 4.66 6.25 26.23 69 LDenR(2) -6.313 7.727 0.892 8.135
(2) versus TL [-20.570, 7.944] [7.071, 8.384]
124 LDenR(2) -5.034 6.577 0.979 4.610
versus FL [-8.468, -1.601] [6.402, 6.752]
LDenR(2) 2.43 ´ 10-4 3.760 0.982 0.149
versus W [1.80 ´ 10-4, 3.20 ´ 10-4] [3.669, 3.851]
Maxilla left 20.86 5.2 6.37 28.86 69 LMaxL -4.137 7.010 0.892 8.073
versus TL [-18.195, 9.921] [6.415, 7.605]
121 LMaxL -2.096 5.911 0.977 4.722
versus FL [-5.604, 1.411] [5.747, 6.075]
LMaxL 2.62 ´ 10-4 3.632 0.983 0.147
versus W [2.00 ´ 10-4, 3.40 ´ 10-4] [3.546, 3.718]
Maxilla right 20.82 5.11 6.44 28.72 69 LMaxR -2.965 6.957 0.866 8.859
versus TL [-18.712, 12.782] [6.291, 7.623]
124 LMaxR -0.922 5.864 0.969 5.410
versus FL [-4.918, 3.074] [5.677, 6.051] 5.410
LMaxR 2.91 ´10-4 3.601 0.980 0.161
versus W [2.20 ´10-4, 3.80 ´10-4] [3.509, 3.693]
Anterior 11.62 2.86 3.39 16.03 66 LCerL -0.172 12.350 0.880 8.544
ceratohyal versusTL [-15.169, 14.824] [11.213, 13.488]
left 122 LCerL -4.349 10.779 0.969 5.599
versus FL [-8.526, -0.172] [10.429, 11.129]
LCerL 1.79 ´10-3 3.710 0.972 0.188
versus W [1.36 ´10-3, 2.35 ´10-3] [3.597, 3.823]
Anterior 11.66 2.84 3.45 16.02 68 LCerR -5.87 ´10-3 12.285 0.882 8.526
ceratohyal versus TL [-14.521, 14.509] [11.181, 13.388]
right 124 LCerR -2.398 10.613 0.970 5.382
versus FL [-6.398, 1.602] [10.277, 10.948]
LCerR 2.17 ´10-3 3.632 0.974 0.182
versus W [1.67 ´10-3, 2.82 ´10-3] [3.525, 3.739]
Cleithrum 14.07 3.76 5.38 20.47 68 LCleiL 22.841 8.550 0.897 7.971
left versus TL [11.251, 34.431] [7.837, 9.262]
123 LCleiL 6.031 8.163 0.983 3.985
versus FL [3.218, 8.844] [7.969, 8.357]
LCleiL 1.25 ´10-3 3.581 0.983 0.144
versus W [1.00 ´10-3, 1.57 ´10-3] [3.496, 3.666]
247
FERNÁNDEZ ET AL.: DIAGNOSTIC ELEMENTS REGRESSIONS FOR ARGENTINE ANCHOVY
Table 2. Continued.
Diagnostic Mean SD Min. Max. N Relationship Parameters [CI 95%inf, CI 95%sup] R2 RMSE
element
a b
Cleithrum 14.13 3.67 5.95 20.36 67 LCleiR 20.772 8.694 0.910 7.423
right versus TL [9.764, 31.781] [8.016, 9.372]
120 LCleiR 7.147 8.076 0.982 3.949
versus FL [4.278, 10.016] [7.878, 8.273]
LCleiR 1.59 ´10-3 3.491 0.986 0.124
versus W [1.30 ´10-3, 1.93 ´10-3] [3.416, 3.567]
Otolith left 3.9 0.55 2.35 4.96 68 LOtoL(1) -9.957 41.914 0.859 9.223
(1) versus TL [-27.033, 7.118] [37.738, 46.089]
104 LOtoL(1) -22.267 38.916 0.907 6.865
versus FL [-31.911, -12.624] [36.469, 41.364]
LOtoL(1) 0.070 4.179 0.903 0.206
versus W [0.048, 0.101] [3.910, 4.449]
Otolith left 2.02 0.27 1.39 2.59 68 LOtoL(2) -4.477 78.052 0.717 13.361
(2) versus TL [-30.045, 21.092] [66.009, 90.095]
108 LOtoL(2) -26.236 76.866 0.838 9.209
versus FL [-39.495, -12.977] [70.359, 83.374]
LOtoL(2) 0.948 4.361 0.820 0.277
versus W [0.717, 1.253] [3.967, 4.755]
Otolith right 3.9 0.57 2.38 5.08 68 LOtoR(1) -4.449 40.468 0.842 10.021
(1) versus TL [-22.143, 13.244] [36.154, 44.782]
104 LOtoR(1) -19.839 38.183 0.903 7.143
versus FL [-29.523, -10.157] [35.727, 40.638]
LOtoR(1) 0.077 4.106 0.899 0.211
versus W [0.053, 0.111] [3.836, 4.376]
Otolith right 2.02 0.27 1.43 2.74 68 LOtoR(2) -0.024 76.251 0.671 14.159
(2) versus TL [-27.748, 27.700] [63.143, 89.359]
106 LOtoR(2) -23.083 75.464 0.809 10.101
versus FL [-37.588, -8.578] [68.337, 82.589]
LOtoR(2) 0.992 4.307 0.793 0.305
versus W [0.733, 1.342] [3.879, 4.736]
diagnostic elements, as well as to estimate the wet
weight of anchovies using bones and otoliths
(Table 2). All relationships were significant (p-
value <0.05). Generally, a strong correlation was
observed between diagnostic elements and
lengths, with R2values exceeding 0.81 for furca
length and 0.67 for total length. The R2values
greater than 0.793 for the regressions with the
wet weight as the response variable were
obtained. The slope bdid not significantly differ
between measurements of the right and left diag-
nostic elements (ANCOVA, p-value >0.05).
Overall, the prediction error (RMSE) was low,
ranging from 0.11 to 14.159, indicating the relia-
bility of the proposed regressions for predicting
anchovy size and weight.
This study contributes to the use of different
diagnostic elements such as otoliths, pectoral and
head bones to estimate the length and weight of
Argentine anchovy, a highly relevant species in
the southwestern Atlantic Ocean and a signifi-
cant prey item for numerous top predators
(Velasco and Castello 2005). The analyzed
length range encompasses the maximum size
recorded for the Patagonian stock (198 mm,
Hansen 2004) and the prey size range observed
in the diet of various important marine predators
in the region, including fish (Koen Alonso et al.
2001; Sánchez 2009), seabirds (Gatto and Yorio
2009; Castillo et al. 2019), and marine mammals
(Koen Alonso et al. 1998; Koen Alonso et al.
2000).
Traditionally, otoliths have been commonly
used in diet studies to estimate length and/or
weight of consumed prey. However, regressions
conducted in this study exhibited lower coeffi-
cients of determination (R2) and higher RMSE for
otoliths compared to other elements. This dis-
crepancy may be attributed to differences in the
otolith length and fish growth relationship
(Laidig and Ralston 1995). Typically, otoliths
exhibit a linear relationship until fish reaches its
maximum size, after which it begins to increase
in thickness (Blacker 1974). In contrast, cranial
bone lengths exhibited stronger relationships,
with R2values ranging from 0.866 to 0.989. This
is because fish length and the length of some
growth bones are constant (Prenda and Granado-
Lorencio 1992; Carss and Elston 1996). More-
over, cranial and pectoral bones are easily recog-
nizable and identifiable at the species level (Pren-
da and Granado-Lorencio 1992) and more resist-
ant to erosion compared to otoliths (Carss and
Elston 1996), which can introduce lower biases in
size calculation (Scharf et al. 1998). For this rea-
son, durable bones like cleithrum and dentary are
preferred. In our study, the best regression esti-
mates were obtained using these two bones.
Our study indicated positive allometric growth
in Patagonian anchovy stock samples, suggesting
that its weight increases more than its predicted
total length. The bparameter estimated in previ-
ous studies for the Patagonian anchovy stock
ranged from 3.21 to 3.35 (Hansen 2004), while it
was estimated from 2.97 to 3.40 for the
Bonaerensis stock (Hansen 1997; Haimovici and
Velasco 2000; Garciarena et al. 2002; Hansen
2004). Differences in these values may be attrib-
uted to various sampling-related factors, such as
sample size, length range, conservation method,
temporal resolution of sampling, sex, and stage
of specimens, as well as variations between
stocks or those from different geographical
regions (Weatherley and Gill 1987; Froese et al.
2011; Kuriakose 2017).
Although this study provides valuable
insights, it should be noted that collection meth-
ods used did not allow for the acquisition of
smallest sizes of Argentine anchovy (<40 mm
furca length), which are particularly challenging
to obtain. However, fortuitous stranding events
of this species in Golfo Nuevo, Chubut, in 1997
and 2000, did allow for the collection of some of
the smallest sizes (2023 pers. comm. A
Gosztonyi). Additionally, it should be noted that
these samples (17 of the 125 individuals) were
preserved in formalin, so caution should be exer-
cised when estimating wet weight based on our
248 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
relationship within this size range, as it may
underestimate individual weight.
These relationships can be valuable in compar-
ative studies of growth, determination of body
condition and calculation of total biomass con-
sumed in diet studies, such as those of Fernández
et al. (2019) and Ibarra et al. (2022), and estima-
tion of energy density in this Patagonian popula-
tion (Ciancio et al. 2020), contributing to the
understanding of the trophic ecology of predators
in one of the world’s most productive oceans.
ACKNOWLEDGMENTS
We express our gratitude to Dr Julio Lancelot-
ti, Dr Atila Gosztonyi, Dr Pablo Yorio, Dr Crist-
ian Marinao, Nestor Ortíz, and the crew of the
ship ‘El Padrino’ in Puerto de Rawson for their
valuable assistance in sample collection. Special
thanks go to Dr Gabriela Escati Peñaloza from
the Instituto Nacional de Investigación y Desar-
rollo Pesquero (INIDEP) for her contributions.
Lastly, we extend our appreciation to Dr Cristian
‘Kily’ Durante for his expertise in image editing,
and the anonymous reviewers to their valuable
comments.
Author contributions
Fernández J. Santiago: conceptualization, data
curation, formal analysis, investigation, method-
ology, roles/writing-original draft, writing-review
and editing. Cynthia Ibarra: conceptualization,
data curation, formal analysis, investigation,
methodology, roles/writing-original draft, writ-
ing-review and editing. Ximena Navoa: formal
analysis, methodology, supervision, roles/writ-
ing-original draft, writing-review and editing.
Javier E. Ciancio: project administration, supervi-
sion, validation, visualization, writing-review and
editing.
REFERENCES
BELLEGGIA M, FIGUEROA DE, SÁNCHEZ F, BRE-
MEC C. 2012. Long-term changes in the spiny
dogfish (Squalus acanthias) trophic role in the
southwestern Atlantic. Hydrobiologia. 684:
57-67. DOI: https://doi.org/10.1007/s10750-
011-0967-y
BLACKER R. 1974. Recent advances in otolith
studies. In: JONES F, editor. Sea Fisheries
Research. New York: Wiley.
CARLANDER KD, SMITH LL. 1945. Some factors
to consider in the choice between standard,
fork, or total lengths in fishery investigations.
Copeia. 1945: 7-12.
CARSS DN, ELSTON DA. 1996. Errors associated
with otter Lutra lutra faecal analysis. II. Esti-
mating prey size distribution from bones
recovered in spraints. J Zool. 238 (2): 319-
332. DOI: https://doi.org/10.1111/j.1469-799
8.1996.tb05397.x
CASTILLO J, YORIO P, GATTO A. 2019. Shared
dietary niche between sexes in Magellanic
Penguins. Austral Ecol. 44: 635-647. DOI:
https://doi.org/10.1111/aec.12706
CIANCIO JE, BARTES S, FERNÁNDEZ S, HARILLO C,
LANCELOTTI J. 2020. Energy density predic-
tors for Argentine anchovy Engraulis anchoi-
ta, a key species of the southwestern Atlantic
Ocean. Trans Am Fish Soc. 149: 204-212.
DOI: https://doi.org/10.1002/tafs.10223
CIANCIO JE, YORIO P, BURATTI C, COLOMBO GÁ,
FRERE E. 2021. Isotopic niche plasticity in a
marine top predator as indicator of a large
marine ecosystem food web status. Ecol Indic.
126: 107687. DOI: https://doi.org/10.1016/j.
ecolind.2021.107687
CRESPO EA, PEDRAZA SN, DANS SL, KOEN ALON-
SO M, REYES LM, GARCÍA NA, COSCARELLA
M, SCHIAVINI ACM. 1997. Direct and indirect
effects of the highseas fisheries on the marine
mammal populations in the northern and cen-
249
FERNÁNDEZ ET AL.: DIAGNOSTIC ELEMENTS REGRESSIONS FOR ARGENTINE ANCHOVY
tral Patagonian coast. J Northw Atl Fish Sci.
22: 189-207. DOI: https://doi.org/10.2960/J.
v22.a15
FERNÁNDEZ SJ, YORIO P, CIANCIO JE. 2019. Diet
composition of expanding breeding popula-
tions of the Magellanic Penguin. Mar Biol
Res. 15: 84-96. DOI: https://doi.org/10.1080/
17451000.2019.1596286
FOX J, FRIENDLY GG, GRAVES S, HEIBERGER R,
MONETTE G, NILSSON H, RIPLEY B, WEISBERG
S, FOX MJ, SUGGESTS M. 2007. The car pack-
age. R Foundation for Statistical Computing.
1109: 1431.
FROESE R. 2006. Cube law, condition factor and
weight-length relationships: history, meta‐
analysis and recommendations. J Appl Ichthy-
ol. 22: 241-253. DOI: https://doi.org/10.1111/
j.1439-0426.2006.00805.x
FROESE R, TSIKLIRAS AC, STERGIOU KI. 2011.
Editorial note on weight-length relations of
fishes. Acta Ichthyol Piscat. 41: 261-263.
DOI: https://doi.org/10.3750/aip2011.41.4.01
GARCIARENA AD, PERROTTA RG, LÓPEZ F. 2002.
Informe sobre el muestreo de desembarque de
anchoíta (Engraulis anchoita) y caballa
(Scomber japonicus) en el puerto de Mar del
Plata: período septiembre 1999-enero 2000,
con algunos comentarios sobre el manejo de
estos recursos. INIDEP Inf Téc. 45. 17 p.
GATTO AJ, YORIO P. 2009. Provisioning of mates
and chicks by Cayenne and Royal Terns:
resource partitioning in northern Patagonia,
Argentina. Emu. 109: 49-55. DOI: https://doi.
org/10.1071/MU08025
GONZÁLEZ-ZEVALLOS D, KUBA L, GOSZTONYI AE.
2010. Estimación de la longitud utilizando
relaciones morfométricas de huesos del crá-
neo, cintura escapular, otolitos y medidas
específicas del cuerpo en Merluccius hubbsi
en aguas patagónicas. Rev Biol Mar Ocea-
nogr. 45 (2): 341-345. DOI: https://doi.org/
10.4067/S0718-19572010000200017
GOSZTONYI AE, KUBA L. 1996. Atlas de los hue-
sos craneales y de la cintura escapular de
250 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
peces costeros patagónicos. Inf Téc Fund Pata-
gon Nat. 4: 29 p.
HAIMOVICI M, VELASCO G. 2000. Relações com-
primento peso de peixes teleósteos marinhos
do sul do Brasil com uma avaliação de diferen-
tes métodos de ajuste. Atlantica. 22: 131-140.
HANSEL HC, DUKE SD, LOFY PT, GRAY GA. 1988.
Use of diagnostic bones to identify and esti-
mate original lengths of ingested prey fishes.
Trans Am Fish Soc. 117: 55-62. DOI:
https://doi.org/10.1577/1548-8659(1988)117<
0055:UODBTI>2.3.CO;2
HANSEN JE. 1997. Muestreo bioestadístico de pes-
cado en el Puerto de Mar del Plata. Anchoíta
(Engraulis anchoita). Período 1986-1990. Inf
Téc INIDEP. 15: 1-24.
HANSEN JE. 2004. Anchoíta (Engraulis anchoi-
ta). In: SÁNCHEZ RP, BEZZI SI, editors. El Mar
Argentino y sus recursos pesqueros. Tomo 4.
Los peces marinos de interés pesquero carac-
terización biológica y evaluación del estado
de explotación. Mar del Plata: Instituto Nacio-
nal de Investigación y Desarrollo Pesquero
(INIDEP). p. 101-115.
HÓDAR J. 1997. The use of regresion equations for
the estimation of prey length and biomass in
diet studies of insectivore vertebrates. Arx
Misc Zool. 20: 1-10.
IBARRA C, MARINAO C, SUÁREZ N, KASINSKY T,
YORIO P. 2022. Patterns of sexual segregation
in the use of trophic resources in breeding
Imperial Cormorants. Mar Biol. 169: 154.
DOI: https://doi.org/10.1007/s00227-022-04
143-7
IBARRA C, MARINAO C, SUÁREZ N, YORIO P. 2018.
Differences between colonies and chick-rear-
ing stages in Imperial Cormorant (Phalacroco-
rax atriceps) diet composition: implications
for trophic studies and monitoring. Wilson J
Ornithol. 130: 224-234. DOI: https://doi.org/
10.1676/16-184.1
KOEN ALONSO M, CRESPO E, GARCÍA N, PEDRAZA
S, MARIOTTI P, VERA BB, MORA N. 2001. Food
habits of Dipturus chilensis (Pisces: Rajidae)
off Patagonia, Argentina. ICES J Mar Sci. 58:
288-297. DOI: https://doi.org/10.1006/jmsc.
2000.1010
KOEN ALONSO M, CRESPO EA, GARCÍA NA,
PEDRAZA SN, COSCARELLA MA. 1998. Diet of
dusky dolphins, Lagenorhynchus obscurus, in
waters off Patagonia, Argentina. Fish Bull. 96:
366-374.
KOEN ALONSO M, CRESPO EA, PEDRAZA SN,
GARCIA NA, COSCARELLA MA. 2000. Food
habits of the South American sea lion, Otaria
flavescens, off Patagonia, Argentina. Fish
Bull. 98: 250-263.
KURIAKOSE S. 2017. Estimation of length weight
relationship in fishes. Course Manual Summer
School on advanced methods for fish stock
assessment and fisheries management lecture
note series no. 2/2017. CMFRI, Kochi, p. 215-
220.
LAIDIG TE, RALSTON S. 1995. The potential use of
otolith characters in identifying larval rockfish
(Sebastes spp.). Fish Bull. 93: 166-171.
LINHART H, ZUCCHINI W. 1986. Model selection.
New York: John Wiley & Sons.
LOIZAGA DE CASTRO R, SAPORITI F, VALES DG, GAR-
CÍA NA, CARDONA L, CRESPO EA. 2016. Feeding
ecology of dusky dolphins Lagenorhynchus
obscurus: evidence from stable isotopes. J
Mammal. 97 (1): 310-320. DOI: https://doi.
org/10.1093/jmammal/gyv180
MARTÍNEZ‐POLANCO MF, BÉAREZ P, JIMÉNEZ‐
ACOSTA M, COOKE RG. 2022. Allometry of
Mexican hogfish (Bodianus diplotaenia) for
predicting the body length of individuals from
two pre‐Columbian sites in the Pearl Island
archipelago (Panama). Int J Osteoarchaeol.
32: 669-681. DOI: https://doi.org/10.1002/oa.
3094
PEREZ COMESAÑA J, CLAVIN P, ARIAS K, RIESTRA
C. 2014. Total length estimation of the Brazil-
ian flathead Percophis brasiliensis, using mor-
phometric relationships of skull, pectoral gir-
dle bones, otoliths and specific body meas-
ures, in Argentine waters. J Appl Ichthyol. 30:
377-380. DOI: https://doi.org/10.1111/jai.12244
PIROTTA V, M CINTOSH R, GRAY R, MARSH H,
LYNCH M. 2022. Marine mammals. In: SMITH
BP, WAUDBY CA, HAMPTON JO, editors.
Wildlife Research in Australia: practical and
applied methods. Clayton South: CSIRO Pub-
lishing. p. 465-477.
PRENDA J, GRANADO-LORENCIO C. 1992. Biomet-
ric analysis of some cyprinid bones of prey
fishes to estimate the original lengths and
weights. Folia Zool. 41: 175-185.
R CORE TEAM. 2021. R: A language and environ-
ment for statistical computing. Vienna: R
Foundation for Statistical Computing.
RAMOS JA, FURNESS RW. 2022. Seabirds as indi-
cators of forage fish stocks. In: RAMOS J,
PEREIRA L, editors. Seabird biodiversity and
human activities. Boca Raton: CRC Press.
RIESTRA CM, PEREZ COMESAÑA JE, ARIAS KA,
TAMINI LL, CHIARAMONTE GE. 2020. Back-
calculation of total length of Argentine
seabass Acanthistius patachonicus using mor-
phometric relationships of bones and meas-
urements of the body. Mar Fish. 33: 69-75.
DOI: https://doi.org/10.47193/mafis.3312020
061804
SÁNCHEZ F. 2009. Alimentación de la merluza
(Merluccius hubbsi) en el Golfo San Jorge y
aguas adyacentes. INIDEP Inf Téc. 75. 21 p.
SCHARF F, YETTER R, SUNNINNER A, JUANES F.
1998. Enhancing diet analyses of piscivorous
fishes in the Northwest Atlantic through iden-
tification and reconstruction of original prey
sizes from ingested remains. Fish Bull. 96:
575-588.
SCHINDELIN J, RUEDEN CT, HINER MC, ELICEIRI
KW. 2015. The ImageJ ecosystem: an open
platform for biomedical image analysis. Mol
Reprod Dev. 82: 518-529. DOI: https://doi.
org/10.1002/mrd.22489
SINOVČIĆ G, FRANIČEVIĆ M, ZORICA B, ČIKEŠ‐
KEČ V. 2004. Length-weight and length-length
relationships for 10 pelagic fish species from
the Adriatic Sea (Croatia). J Appl Ichthyol. 20:
251
FERNÁNDEZ ET AL.: DIAGNOSTIC ELEMENTS REGRESSIONS FOR ARGENTINE ANCHOVY
156-158. DOI: https://doi.org/10.1046/j.1439-
0426.2003.00519.x
SOKAL R, ROHLF F. 1987. Biostatistics. 2nd ed.
New York: Freeman. 301 p.
TAPELLA F, LOVRICH GA. 2006. Morphological
differences between ‘subrugosa’ and ‘gre-
garia’ morphs of adult Munida (Decapoda:
Anomura: Galatheidae) from the Beagle
Channel, southern South America. J Mar Biol
Assoc UK. 86: 1149-1155. DOI: https://doi.
org/10.1017/S0025315406014123
VELARDE E, EZCURRA E, ANDERSON DW. 2013.
Seabird diets provide early warning of sardine
fishery declines in the Gulf of California. Sci
Rep. 3: 1-6. DOI: https://doi.org/10.1038/srep
01332
VELASCO G, CASTELLO JP. 2005. An ecotrophic
model of southern Brazil continental shelf and
fisheries scenarios for Engraulis anchoita
(Pisces, Engraulididae). Atlantica. 27: 59-68.
WEATHERLEY A, GILL H. 1987. The biology of
fish growth. Orlando: Academic Press.
252 MARINE AND FISHERY SCIENCES 37 (1): 241-252 (2024)
