ICES Journal of
Marine Science
ICES Journal of Marine Science (2012), 69(10), 1840–1849. doi:10.1093/icesjms/fss157
Assessment of goose barnacle (Pollicipes pollicipes Gmelin, 1789)
stocks in management plans: design of a sampling program based
on the harvesters’ experience
J. M. Parada1 *, R. Outeiral 2, E. Iglesias 3, and J. Molares4
1
Centro de Investigacións Mariñas, Consellerı́a do Mar, Pedras de Corón s/n, Box 13, 36620, Vilanova de Arousa, Galicia, Spain
Confrarı́a de pescadores de A Guardia. Baixo Muro, 32. 36780 A Guardia, Galicia, Spain
3
Confrarı́a de pescadores de Baiona. Alférez Barreiro, s/n. Casa do mar, 28. 36300 Baiona, Galicia, Spain
4
Dirección Xeral de Desenvolvemento Pesqueiro. Consellerı́a do Mar. Xunta de Gailica. R/ Irmandiños, s/n. 15701. Santiago de Compostela, Galicia, Spain
2
*Corresponding Author: tel: +34 886 206 348; fax: +34 886 206 372; e-mail: [email protected]
Parada, J. M., Outeiral, R., Iglesias, E., and Molares, J. 2012. Assessment of goose barnacle (Pollicipes pollicipes Gmelin, 1789) stocks in
management plans: design of a sampling program based on the harvesters’ experience. – ICES Journal of Marine Science, 69: 1840 – 1849.
Received 26 April 2012; accepted 22 August 2012.
Management plans of coastal marine resources require a wealth of information on socioeconomic topics, harvesting activities, population
dynamics, and stock status. Moreover, the information provided by technical experts must take into account the needs of the managers.
It must also adapt to schedules to serve a useful purpose. In many cases, the methodologies used by research teams are not directly
applicable as they may be too complicated, aimed at specific objectives related to basic scientific work, or too costly to apply to
long-term monitoring of extensive shellfish beds. Also, rocky coastlines exposed to heavy wave action preclude the use of sampling techniques that involve time-consuming data collection. This paper proposes a quick and simple methodology for gathering data in the field,
based on the knowledge of the harvesters, to obtain stock assessments in keeping with their information needs. This methodology uses
coverage percentage as an abundance index and weighting factor for the biometric information gathered from 50 specimens in each
sampling. The sampling design uses the knowledge of the harvesters to define homogeneous strata. The results are in agreement
with both the scientific-technical knowledge and the harvesters’ knowledge of the populations being analysed.
Keywords: Assessment, goose barnacle, harvesting, Pollicipes pollicipes, sampling methodology, stock.
Introduction
The goose barnacle, Pollicipes pollicipes, is harvested as a shellfishery resource in a number of different regions on the
Atlantic Rim from Brittany to southern Portugal, as well as in
some North African countries such as Morocco (Barnes, 1996).
Other goose barnacle species are harvested in Canada and
Peru. In Galicia (northwest Spain) this fishery is of great significance both socially and economically. From 2003 to 2011 it
reached a mean first sale price of 29 E/kg and a total mean
value of 11 million euros annually (source: www.pescadegalicia.
com). Goose barnacle extraction in Galicia is performed by more
than 1500 harvesters who are grouped into 30 fishers’ guilds
called “cofradı́as de pescadores” (Spanish organizations of artisanal fishers) generally located in small coastal towns where artisanal fishing is the main source of economic activity. The
management of this fishery is based on a system similar to informative co-management (Sen and Nielsen, 1996), which
takes advantage of territorial user rights for fishing (TURFs)
whereby the responsibility is shared by the harvesters (through
their guilds) and the fishery administration (Molares and
Freire, 2003). One of the main characteristics of the extraction
of this species in Galicia is its management through annual
exploitation plans designed by the harvesters associated in
“cofradı́as”, and validated by the Administration if the proposal
falls inside the limits of the legal framework. The exploitation
plans allow for the development of adaptive management that
takes into account stock characteristics and market conditions
at all times (Parada and Molares, 2012) and they have proved
to be a tool that can be applied to other goose barnacle fisheries
(Jacinto et al., 2010).
# 2012 International Council for the Exploration of the Sea. Published by Oxford University Press. All rights reserved.
For Permissions, please email: [email protected]
1841
Assessment of goose barnacle stocks in management plans
Most of the “cofradı́as” that exploit the goose barnacle have a
technical expert (generally a person with a degree in biology or
marine sciences) whose services are financed by the fishery administration. Among their many functions, these experts are in charge
of collecting catch and fishing effort data, carrying out stock
assessments and writing up reports on the technical aspects of
the exploitation plans. Their job, however, is not focused exclusively on the exploitation of the goose barnacle. They must also
cover other S-fisheries (Orensanz, et al, 2005) such as seaweeds,
sea urchins, and bivalves inhabiting the sand beds. Therefore,
since only one technical expert has to attend to the exploitation
of several species, the control of the fishery is limited to monitoring the catches by means of “control points” (generally attended
to by a surveillance service) strategically located in the harvesting
zones. The data collected at these control points include the
amount extracted per person, effort exerted, and catch sizes.
However, while the information provided by the control points
is valuable to management (Jacinto et al., 2010), its assessment
should not be based exclusively on these data. Drawing up and
implementing exploitation plans in quotas require independent
catch information such as population dynamics, stock abundance,
and the stock structure of the exploited species (FAO, 2005). The
availability of data for population estimations in terms of coverage,
size structures, reproductive states, and abundance will help
provide a foundation on which to base decisions in matters
related to the rotation of extraction zones, evaluation of impact,
interaction with other fisheries, mortalities, etc.
The scientific knowledge applied to resource management has
not always taken into account the actual needs or the previous
knowledge of the communities that exploit these resources
(Lauzier, 1999; Garcı́a Allut, 2000). The fishery administration
also requires information on populations to evaluate the exploitation plans and to understand the situation along the entire
coast. However, owing to the lack of a standardized methodology
that can adapt to the possibilities and the means available to the
technical experts, it is difficult to obtain comparable information.
Moreover, most of the methodologies for goose barnacle population
assessment require several people and are very time-consuming.
These methods become difficult to carry on year after year, particularly by only one person trying to assess an extensive barnacle
bed. Due to the difficulties in obtaining data and maintaining
long-term stock assessment programs, informative indicators
based on fisheries statistics and fishers’ behaviour are currently
used (Freire and Garcı́a Allut, 2000). However, management
decisions could be better supported by handling both sources of
information, stock assessment and fisheries statistics.
The objective of this paper is to propose a quick and simple
methodology that can be applied to the study of goose barnacle
populations; a methodology that takes the harvesterś knowledge
into account and provides, in a sustainable way, essential information needed to manage the exploitation of the resource.
barnacles, to compare it with the values found in previous years,
and then relate it to the catches. In Galicia, both the harvesters
and the market itself make a distinction between two large
groups of goose barnacles on the basis of their morphology: elongated goose barnacles and standard goose barnacles (Figure 1).
The former have a lower market value, and some attempts have
been made to use alternative commercial methods, including
canning. So, harvesters and managers are not interested only in
knowing the abundance of the population as a whole, but also
the abundances of each of the two typologies. Therefore, assessments must refer to both the population as a whole and to the
stocks of elongated and standard goose barnacles separately.
They must also allow for the application of techniques for the
study of population dynamics. In addition, to give a wider characterization of the stock dynamic to improve the stock assessment,
they must be able to report on the reproductive state and recruitment of the population (Haddon, 2011).
Sampling design
A sampling strategy that meets the requirements described above
was designed (Figure 2). To evaluate its suitability, the strategy
was implemented on the goose barnacle beds that were accessible
from the coast and exploited by the Cofradı́as of A Guardia and
Baiona, located in southwestern Galicia. (Figure 3). The sampling
stations were organized on the basis of a stratified design in which
each stratum included sections of the coast with homogeneous
characteristics. In view of the relationship between the characteristics of the coast and the population parameters, (Borja et al.,
2006a), the identification of the strata was based on the sections
identified, on aerial photographs, by the local harvesters in a previous study conducted by Centro Tecnológico del Mar (CETMAR)
(unpublished data) according to the commercial classification of
the goose barnacles collected in each one. In this way, five strata
(types of coast) were identified in keeping with the quality of the
goose barnacle: very high, high, average, low, very low.
According to their experience, harvesters based this classification
on the general shape of goose barnacles. In the Cofradı́a of
Baiona, the very high quality stratum could not be sampled
owing to the impossibility of gaining safe access to the area
during the study period (winter–spring of 2011).
Methods
Requirements
The methodology proposed here considers that the collection and
processing of the samples must be quick and simple and that these
processes must be able to be carried out by a single person.
However, due to safety reasons, working on the rocky coast
requires the collaboration of another person for support. The
data obtained must be able to estimate the abundance of goose
Figure 1. Goose barnacle specimens having an elongated typology
and a wrinkled peduncle (a), and standard specimens with a smooth
peduncle (b).
1842
J. M Parada et al.
Figure 2. Sampling design. Tasks in frames with rounded corners are developed in harvesters meetings. Tasks in regular frames are carried out
at office. Tasks in shaded frames refer to field works. Dashed tasks have not been tackled in this paper.
Figure 3. Goose barnacle beds in the fishers’ guilds cited in this
study.
Initially, a number of samples proportional to the relationship
between the total length of each type of coast and the total length
of the coast with the presence of goose barnacles were determined
in each stratum. Nineteen stations (57 samples) were established in
Baiona and 18 in A Guarda. However, data processing was carried
out using the methodology pertaining to the optimized stratification design (Murray et al., 2006) in which the number of samples
in each stratum is not proportional to the stratum size, but may
increase in the strata with the greatest variance to improve the
accuracy of the estimations (Krebs, 1998). In optimized stratification, it is necessary to know the proportion of the sampled surface
area to the total surface area of each stratum. To estimate the area
of each stratum, the sum of the lengths of each section of coast
belonging to the stratum was multiplied by the mean width of
the goose barnacle belt (Borja et al., 2006a). In each sampling
station, samples were collected at three points at regular intervals
along a line that went from the goose barnacle belt (between about
1 and 3 m above the MLW, depending on the exposure of the
coast), in a transversal direction, to the intertidal strip. This
allowed for the inclusion of any possible variability due to tidal
level and, at the same time, for the analysis of the effect of this
factor on the variables studied. Cracks and cavities were avoided
during sample collection.
Each sampling consisted of a photograph taken of an area
measuring 50 × 50 cm and the collection of between 30 and 50
specimens, preferably from the same cluster within the area photographed. In cases where there were not enough specimens in the
photographed square, samples were collected out of the square
within the same tidal level. The photographs and specimens collected were processed in the lab. Once in the laboratory, a digital
layer with a template of 100 regularly distributed points was superimposed on each photograph (Foster et al., 1991). In this way it
was possible to calculate the coverage as an abundance index on
the basis of the percentage of points in the template that contacted
any goose barnacle. The specimens collected were separated from
the clusters and then measured to determine their total height
(TH) and capitulum base diameter (DBC) interpreted as the
distance between the base of the rostrum and the apex of the subcarina (Figure 4). Although the size that best represents growth is
1843
Assessment of goose barnacle stocks in management plans
Table 1. Capitulum base diameter – total height relationship data
series used in the validation of the discrimination function of the
two goose barnacle typologies elongated and standard.
Fishers’ guild
A Guarda
A Guarda
Corme
Corme
A Guarda, Baiona and Bueu
Typology
elongated
standard
elongated
standard
elongated
ID
AG_E
AG_S
CR_E
CR_S
CT_E
n
80
80
38
47
808
ID ¼ Series identifier, n ¼ number of specimens. Series CT_E comes from a
fisheries management study on the goose barnacle conducted by CETMAR
and the fishers’ guilds of origin with data ranging from 2008 to 2010
(unpublished data).
Figure 4. The most commonly used biometries in goose barnacle
studies and capitulum plates used as a reference. TH ¼ total height,
RC ¼ rostrum– carina distance, DBC ¼ capitulum base diameter,
R ¼ rostrum, C ¼ carina, SC ¼ subcarina.
the rostrum - carina distance (RC) length (Cruz, 1993), the DBC
was adopted here since it is the size established by the Fishery
Administration in Galicia to determine the minimum catch size.
Moreover, the DBC is an unequivocal size and very little training
is needed to be able to measure the specimen correctly (Parada
et al., in press). DBC is also related to the diameter of the peduncle,
which is, in turn, related to the commercial quality of the goose
barnacle (Molares et al., 1987). Where needed for comparison
with other research, the results of the DBC were converted to
RC by means of the relationship described by Parada et al. (in
press).
RC = 1.0666 DBC − 2.2875
The DBC data were used to calculate the size structure with
1-mm size– class ranges and to differentiate between the stock of
commercial and non-commercial-sized specimens. The DBC –
TH relation was employed to differentiate between the specimens
with an elongated typology from those having a standard typology,
as detailed below, to meet the information needs of managers and
harvesters alike (Lauzier, 1999). Therefore, total coverage in each
station was corrected by the percentages of commercial and noncommercial specimens to be able to estimate the coverage of these
categories by station:
CC = TC · PC
where CC represents commercial coverage, TC, total coverage, and
PC, percentage of commercial specimens.
At the same time, the total coverage of each station was corrected with the percentages of elongated and standard type individuals to estimate the coverage of each typology by station.
Adult specimens were examined to determine their reproductive state (RS) and recruitment index (RI). The RS of each sample
was defined as the percentage of adult specimens with ovisacs in
the capitulum cavity (Molares, 1993). On the other hand, the RI
was defined as the percentage of adult individuals with at least
one recruit (DBC , 2.01 mm) recently attached by its peduncle.
Adult individuals were considered to be those with a DBC equal
to or greater than 13 mm (Parada et al. in press).
Morphological definition of elongated and standard
typologies
The distinction between the two types of goose barnacles, elongated and standard, is based on the subjective interpretation of
the harvesters when they select their catches to be sold.
However, in the study of the stocks of these two typologies, it is
necessary to use a procedure that will enable them to be differentiated on the basis of objective biometric parameters. According to
the harvesters, elongated specimens have a longer and thinner
peduncle that the standard type of goose barnacle in relation to
capitulum size (Figure 1). The relationship between the DBC
and TH of the three samples taken in one of the stations having
the broadest size range was studied. The station was located in a
zone called “As Maulas” (42806.166′ N 8853.909′ W), within the
goose barnacle beds pertaining to the Cofradı́a of Baiona. The
339 specimens measured fresh were grouped into 5-mm size
classes according to their DBC. On the basis of the mean TH of
the five size classes obtained, the following type of sigmoid function was calculated
P y = A1 + (A2 − A1 )/1 + x/x0
where A1, A2, x0 and p are constants. The function was used to
mark the border between the elongated and standard typologies.
In this way, specimens with a TH higher than the function corresponding to their DBC were considered elongated, while the rest of
the specimens were classified as standard. The function was validated using three series of DBC/TH data obtained from 1053
goose barnacles previously classified as elongated or standard by
harvesters belonging to four fishers’ guilds (Table 1). Parada
et al. (in press) found that the DBC values did not vary after the
frozen specimens were thawed, whereas the TH values did. For
possible future applications of this curve using the samples that
were frozen, a second curve was obtained for thawed individuals,
fitting the former curve by using the same DBC data and correcting the values corresponding to the TH by means of function
TH* ¼ 1.12 TH – 1.5719, where TH* is the TH of the thawed specimens and TH is the measurement of the fresh specimens (Parada
et al., in press).
Calculation procedure
The mean, variance, and standard error of the total coverage of
commercial and non-commercial size stocks and of the two
typologies, elongated and standard, were calculated following the
calculation procedure used for small samplings of contagious
1844
J. M Parada et al.
Figure 5. Data collected and type of results obtained in each sampling (frames with rounded corners). The shaded squares indicate the type
of estimation made in each stratum.
distributed populations (Elliot, 1977). The size structure for each
stratum was obtained by applying the same procedure to each
size class, but based on the mean and variance of the percentage
of the same size class in all the samplings weighted with the
total coverage. The weighted mean with coverage was estimated
as follows:
M=
n
i=1
n
mi Ci / i=1 Ci
where M is the weighted mean of the percentage of specimens in a
size class, n is the number of samplings, mi is the percentage of specimens in this size class in sampling i, and Ci is the total coverage
in sampling i. The variance estimator (Var) used for the weighted
mean was the one put forth by Thompson (1992):
Var =
n
i=1
Ci (mi − M )2 /(n − 1)
n
i=1
Calculations were performed with the ARouSA application, available at http://sites.google.com/site/arousa09.
Figure 5 presents a diagram of the different data processing
approaches and the type of results yielded. The methodology
was validated by applying the Chi-square test to assess the relationship between the percentage of elongated-type goose barnacles and
total coverage with coast types and tidal level. The identification of
cohorts was carried out by means of the program FISAT.
Moreover, in the case of the Cofradı́a of A Guarda, where data
have been available since 2004, the estimations of coverage were
related to the catches obtained one and two years after each
evaluation. This relationship was established by applying the log
function used by the technical experts at the fishers’ guilds to
draft the exploitation plans of marine resources (Parada &
Molares, 2012):
Ci
The mean and variance of the RS and RI for each stratum were
computed according to the same weighting procedure used with
coverage. Arcsine transformation was previously applied to all percentage data (Krebs, 1998). The values corresponding to each of
the two exploitation areas studied in their entirety (A Guardia
and Baiona) were calculated on the basis of the results obtained
for each stratum (coast with goose barnacles classified into categories of very high, high, average, low and very low) following
the calculation procedure for stratified samplings described by
Krebs (1998). All calculations were carried out for the total stock
and the two typologies, elongated and standard, as well as for
the entire coast and each of the three tidal levels sampled.
y = bLn(x) + a
where a and b are constants, y represents the catches, and x the
abundance indicator estimated in the samplings, in this case,
total coverage.
Results
Definition of standard and elongated typologies
The sigmoid function used as a threshold separating the two typologies, elongated and standard, (Figure 6a) corresponded to the
following function:
y = 62.3122 + (11.4815 − 62.3122)/(1 + (x/12.216)2.8009 )
1845
Assessment of goose barnacle stocks in management plans
Table 2. Percentage of specimens in each size class that
maintained the typology identified by the harvesters after the
application of the discrimination function.
DBC
[4, 6)
[6, 8)
[8, 10)
[10, 12)
[12, 14)
[14, 16)
[16, 18)
[18, 20)
[20, 22)
[22, 24)
[24, 26)
%E
59.2
68.6
74.8
72
77.8
93.8
100
100
100
–
–
%S
–
100
–
100
100
100
79.4
86
84.6
100
100
DBC ¼ Capitulum base diameter, %E ¼ percentage of elongated specimens
that kept the elongated classification, %S ¼ percentage of standard
specimens that retained the standard classification.
classification with that of the harvesters. In the 4 –6 mm DBC
size class, this percentage was only 59.2% (Table 2).
Stock assessment and analysis
Figure 6. Fit of the sigmoid curve separating the elongated and
standard typologies (a). Validation of the curve with data from the
goose barnacles identified as elongated and standard in A Guarda
and Corme (b), and with data from goose barnacles considered to be
elongated by the harvesters from the Cofradı́as of A Guarda, Baiona
and Bueu, taken from the study conducted by CETMAR
(unpublished data) (c). See series identifier in Table 1.
where y is the TH above which the specimens having a specific
DBC (x) are considered to be of the elongated type. In the case
of thawed goose barnacles, the curve corresponded to the following function:
y = 67.8055 + (10.9002 − 67.8055)/(1 + (x/12.216)2.8008 ).
After applying the sigmoid function to the DBC values, most of
the specimens were classified into the same typology as designated
by the harvesters (Figure 6b and c). In six out of the nine size
classes with data, all of the specimens identified as standard by
the harvesters were recognized as standard by the sigmoid function. In the remaining three size classes, identification coincided
in over 79% of the specimens. In the case of the elongated specimens, over 90% having a DBC greater than 13.99 mm were
found to belong to this category. However, the smaller in size
the specimens were, the fewer number of specimens coincided in
The proposed methodology allowed taking data in 3 –4 stations
(9–12 samples) per day, including the localization and identification of each station. Based on this methodology, it was possible
to estimate the coverage and percentage of elongated specimens
in the population as a whole including both commercial and noncommercial specimens. The sampling design was able to yield this
information as well as data related to the RI and RS indices for
both the entire goose barnacle belt and the tidal levels from
which the samples were collected. Although these results were
obtained for each type of coast (stratum), table 3 shows the findings for the entire coast (stratified analysis) in A Guarda. The size
structures for the total population and for the two typologies of
goose barnacle were also obtained (Figure 7).
The mean goose barnacle percentage coverage in A Guarda was
15%, while the coverage of standard type commercial size specimens was only 1.2%. The coverage of the elongated type specimens
accounted for 17.6% of the population. However, the percentage
of specimens belonging to the elongated typology in all of the
strata sampled in the Cofradı́a of Baiona reached a mean value
of 31.6%. A Chi-square test of independence revealed that
the percentage of elongated individuals was not homogeneous
(p ¼ 0.045) between the two fishers’ guilds. Although the highest
percentages of elongated type specimens seem to be associated
with the goose barnacle beds of Baiona, it must also be considered
that in this fishers’ guild, the high-quality stratum was not sampled
(Table 4).
The results of the Cofradı́a of A Guardia suggest that the percentage of elongated type specimens is greater, the lower the
tidal level. However, a Chi-square test of independence indicated
that no significant link exists (p ¼ 0.089 in A Guarda, and p ¼
0.487 in Baiona) between the percentage of elongated type goose
barnacles and the tidal level in either of the fishers’ guilds
(Table 4). In contrast, a significant relationship (p ¼ 0.008 in A
Guarda, and p ¼ 1.80E-06 in Baiona) was found in the two beds
between the percentage of elongated goose barnacles and the
type of coast. The elongated goose barnacles were associated
with coasts having low and very low qualities of goose barnacle,
1846
J. M Parada et al.
Table 3. Results of mean coverage, percentage of elongated type goose barnacles, recruitment index (RI) and reproductive state (RS) on
the coast as a whole and in the three tidal levels of A Guarda.
Total coverage
Coverage standard type
Coverage elongated type
% elongated
RI
RS
Non- commercial
13.6 (1.7)
11.1 (1.3)
2.6 (0.3)
17.7 (0.2)
–
–
Commercial
1.4 (0)
1.2 (0)
0.2 (0)
10.2 (0.1)
–
–
Total
15.0 (1.9)
12.2 (1.4)
2.8 (0.3)
17.6 (0.2)
–
–
>12.99 mm
–
–
–
–
16.8 (0.4)
39.8 (0.3)
In addition to the mean values, the variance is given in parenthesis.
Figure 7. Size structure of total population and both typologies of
goose barnacles in A Guarda. The dotted line indicates the minimum
commercial size.
whereas the standard types were associated with coasts having
mean, high and very high quality goose barnacles. The total population coverage was homogeneous (p ¼ 0.412) when the beds from
the two areas were compared. In both A Guardia and Baiona, the
coverage may be considered similar in the three tidal levels (p ¼
0.916 in A Guarda, and p ¼ 0.956 in Baiona) as well as on the different types of coast (p ¼ 0.098 in A Guarda, and p ¼ 0.132 in
Baiona) (Table 5).
The size structures obtained were successfully used to identify
year classes. In the beds pertaining to the Cofradı́a of Baiona
two age groups were identified: one with a mean size (DBC) of
6.88 mm (SD ¼ 2.17) (corresponding to an RC of 4.97 mm) and
the other with a mean size of 14.55 mm (SD ¼ 2.85) (RC ¼
13.23 mm). The former accounted for 26% of the total population
and the latter 74%. The size structure of the total population of A
Guardia allowed for the identification of two-year classes with
mean sizes of 4.86 mm (SD ¼ 1.51) (RC ¼ 2.90 mm) and
12.03 mm (SD ¼ 3.19) (RC ¼ 10.54 mm). The former comprised
32% of the population and the latter, 68% (Figure 8). The mean
sizes of the year classes from Baiona corresponded to an RC of
4.97 and 13.23 mm.
The estimations of total coverage in A Guardia did not exhibit a
good fit with catch per person and day for the year following the samplings (r 2 ¼ 0.143). They were, however, coherent
(r 2 ¼ 0.895) when compared to the catch data two years after
the estimation (Figure 9).
Discussion
To carry out a technical assessment of the exploitation of marine
resources it is necessary to have a wealth of data on the biology
of the species and the status of the stocks (Caddy and Defeo,
2003). However, in many cases, sufficient resources are not available to manage the large extent of the beds and number of species.
Methodologies used by research groups focusing on very specific
goals are not always directly applicable to this field, which is
more interested in monitoring populations on a middle- or longterm basis, with broader objectives and less time to obtain results.
However, different measures implemented by management, such
as the rotation of exploitation areas, extraction dates in keeping
with market demand, and the possibility of selecting catch
quality depending on the end market, make it absolutely essential
for assessors to know not only the general stock status, but also the
status of specific categories related to this stock.
The technique used to estimate coverage based on photographs
taken in the field and later processed in the lab is quick, simple,
and non-destructive. The speed with which data are recorded is essential when working in the rocky intertidal zones of wave-exposed
coasts (Murray et al., 2006). Several authors have tried different
methods which, while maintaining accuracy, reduce the amount
of time sample collectors must remain on the rocky coast to
collect the data from each sampling point. “Vertical point quadrant” (Johnston, 1957), also named as “point contact procedure”
(Murray et al., 2006), has been the most-used method for
decades. This sampling method provides the most consistent
results and it is the most appropriate for the study of sessile organisms living in the rocky intertidal zone. Foster et al. (1991) applied
photography to the point intercept technique. These authors
pointed out the disadvantages of making calculations based on
photographs rather than calculation in the field with regard to
species identification in synecological studies. This, however,
would not be applicable when working with a single, easily identifiable species. On the other hand, they underlined the great advantages of achieving higher accuracy based on the possibility of using
a larger number of points when working from a photograph. They
also pointed out that for calculation, more time is spent in the lab
than on the rocky coast. The most commonly used technique to
calculate the abundance or biomass indices of the goose barnacle
is to remove the barnacles from a known surface area by scraping.
Borja et al. (2006a) and de la Hoz and Garcı́a (1993) scraped areas
measuring 30 × 30 cm and 25 × 80 cm, respectively. In our
opinion, however, the application of this technique on the
exposed rocky coast is very difficult and very time-consuming.
Another complex aspect of the study of the communities living
on the rocky coast is the estimation of the area occupied by the
population under study (Murray et al., 2006). This is even more
critical when attempting to carry out stock assessments to extrapolate the biomass or individuals per unit area to absolute terms such
as number of individuals or total biomass available. The use of
coverage as a relative abundance index avoids this problem, but
the application of optimized stratification requires that the
1847
Assessment of goose barnacle stocks in management plans
Table 4. Results of the Chi-square test of independence to find associations between goose barnacle typologies, coast types and tidal
levels.
Fishers’ guild
A Guarda
Baiona
A Guarda
A Guarda
A Guarda
Baiona
Baiona
Baiona
A Guarda
A Guarda
A Guarda
A Guarda
A Guarda
Baiona
Baiona
Baiona
Baiona
Type of coast
all
all
all
all
all
all
all
all
very low
low
average
high
very high
very low
low
average
high
Tidal level
all
all
high
average
low
high
average
low
all
all
all
all
all
all
all
all
all
Mean % Elongated
17.6
31.6
12.0
14.6
24.1
26.6
35.9
29.8
20.0
24.7
14.6
5.5
12.7
41.8
44.9
27.8
7.0
Associated typology
standard
elongated
–
–
–
–
–
–
elongated
elongated
standard
standard
standard
elongated
elongated
standard
standard
Chi 2
–
4.02
–
–
4.85
–
–
1.44
–
–
–
–
13.76
–
–
–
29.45
p
–
0.045*
–
–
0.089 n.s.
–
–
0.487 n.s.
–
–
–
–
0.008**
–
–
–
1.80E – 06**
When the associations were significant (*) or highly significant (**) the typology related to each case was indicated. p ¼ probability of the Chi-square test;
n.s. ¼ non-significant associations at a confidence level of 95%. The Yates correction was applied where necessary.
Table 5. Results of the Chi-square test of independence to find
associations between total coverage, coast types and tidal levels.
Fishers’
guild
A Guarda
Baiona
A Guarda
A Guarda
A Guarda
Baiona
Baiona
Baiona
A Guarda
A Guarda
A Guarda
A Guarda
A Guarda
Baiona
Baiona
Baiona
Baiona
Type of
coast
all
all
all
all
all
all
all
all
very low
low
average
high
very high
very low
low
average
high
Tidal
level
all
all
high
average
low
high
average
low
all
all
all
all
all
all
all
all
all
Mean %
Coverage
15.0
19.9
14.3
16.4
14.5
20.1
20.7
18.8
15.9
22.7
15.0
9.0
10.6
21.7
19.3
23.5
10.2
Chi 2
–
0.67
–
–
0.17
–
–
0.1
–
–
–
–
7.8
–
–
–
5.61
p
–
0.412
–
–
0.916
–
–
0.956
–
–
–
–
0.098
–
–
–
0.132
n.s.
n.s.
n.s.
n.s.
n.s.
p ¼ probability of the Chi-square test, n.s. ¼ non-significant association at a
confidence level of 95%. The Yates correction was applied where necessary.
sampled area be related to the total surface area (Krebs, 1998). In
order to simplify the calculation of the rocky coast area by simply
multiplying the length of the coast by the width of the goose barnacle belt, as suggested in this paper and in others (Borja et al.,
2006a), it is necessary to ignore the great heterogeneity of this
medium and consider it to be flat. Due to this simplification,
added to the fact that areas with cracks and cavities were not
included in the sampling, it is impossible to use the biomass estimations as absolute values since they always underestimate the real
biomass. They do, however, offer the possibility of being used as
relative values, which is very useful in comparative analyses and
monitoring the historical series.
The adoption of an objective method to distinguish between
elongated and standard type specimens allows the technical
Figure 8. Identification of the cohorts based on the size structure of
total stocks in Baiona and A Guarda.
experts to apply specific monitoring procedures for each one.
Although the application of the function of typology designation
is biased in the smaller-sized specimens, the results for adults are
very similar to those obtained subjectively by harvesters. In any
case, the results obtained with this function are acceptable, since
even the harvesters find it hard to differentiate between the two
typologies in juvenile specimens. In any event, the greatest importance of these curves lies in their ability to objectively distinguish
the elongated typology from among the marketable specimens,
i.e. the largest sized individuals. The use of the types of coast in
terms of quality as a stratification criterion proved to be suitable
since the results showed that the typologies are related to coast
type (Table 4). Borja et al. (2006b) did not find any significant correlations between total goose barnacle length and the degree of
wave exposure. This could indicate that the morphology of the
goose barnacle, and therefore its commercial quality (Molares
et al., 1987), do not depend exclusively on the degree of wave exposure. Cruz (1993) reported other possible factors such as light
and availability of food, but these aspects have not yet been
1848
Figure 9. Fit of the total coverage estimated for catches per person
and day: 1 year (white squares and dash line) and 2 years (black
circles and solid line) after the estimation. Coverage data from 2004,
2005, 2006, 2008, 2009, and 2010 in A Guarda.
studied. The association between the elongated typology and the
coasts with very low and low quality goose barnacles shows the coherence of the results and validates the methodology used. Hence,
the harvesters’ knowledge of the classification of the different
stretches of coast is a valuable asset when designing sampling programs. The lack of association between goose barnacle typologies
and tidal level (Table 4) is in keeping with the absence of significant correlations between the total length of goose barnacles and
the tidal level described by Borja et al. (2006b). However, the
tidal level is, in fact, related to aspects such as growth and recruitment (Cruz et al., 2010). Therefore it would be of interest for the
sampling design to include a function that would allow the results
to be differentiated between different tidal levels. Moreover, the
availability of the samples in terms of tidal levels will make the information obtained more useful to the organization dedicated to
the exploitation of this resource.
Most sedentary benthic invertebrates, such as goose barnacles,
are structured as metapopulations. Each local population is replenished by the pool of larvae from one or more local populations,
depending on the degree of connectivity and dispersal distance.
In general, these local populations present high variability in their
stock-recruitment relationship, whether they act as “source” or
“sinks” (Shepherd and Brown, 1993). The RS and the RI are used
to characterize the population in one of these two types. In
a source population, the annual recruitment occurs at high
density every year. In a sink population, new age classes are recruited
only in occasionally good years. The exploitation rate should be
different if the population acts as a source or sink. The proposed
methodology allows a quick knowledge of these aspects to be incorporated into the management decisions.
Coverage results like the ones presented in Table 3 in terms of the
two typologies and three tidal levels for each type of coast offer
invaluable information for the application of management tools
such as the rotation of harvesting zones. Although the lack of
association between coverage and tidal level or coast type (Table 5)
may appear to contradict the results reported in other works
(Borja et al., 2006b), this would be expected, taking into account
that the zones studied here are subject to intense exploitation.
The higher percentage of elongated specimens found in Baiona
might be related to some defect in the typology distinction curve of
juveniles. However, the size frequencies showed that the percentage of individuals measuring less than 13 mm was greater in A
J. M Parada et al.
Guarda than in Baiona, so that, in any case, the typology distinction model would favor the designation of more elongated specimens in A Guarda than in Baiona. On the other hand, given the
association between the very high quality type coast and a lower
percentage of elongated specimens, these results will probably
change when samples from this type of coast are available from
the Cofradı́a of Baiona.
The size structures obtained after weighting with coverage
(Figure 8) have made it possible to identify year classes. The
mean size of year classes identified in Baiona (4.97 and
13.23 mm RC) resemble the age group between 0 and 1 year
(5–12.5 y, 12.5– 20.5 mm, respectively) used by Jacinto et al.
(2011), and they are in agreement with the mean sizes reported
by Cruz et al. (2010) 3 and 12 months after scraping (4.6 and
15.7 mm, respectively). However the mean sizes in A Guarda
(2.90 and 10.54 mm RC) were slightly lower. Nevertheless, the
sole purpose of the results obtained here is to validate the sampling
methodology. For the definitive identification of year classes,
several histograms, done consecutively, will be needed.
The coverage presented a better relationship with the catches
recorded two years after the assessment than with those obtained
one year later (Figure 9). This would seem logical, if we consider
that, according to the data of Cruz et al. (2010), this species
would reach a commercial size of 15 mm DBC (13.7 RC) in the
second year (year class I). Given the fact that the harvesters
work more often during a particular tidal level than others, it
can be assumed that with the availability of coverage data specific
to the different tidal levels, the relationship between the coverage
of a particular level and catches will have a better fit. Our results
show that the use of coverage as an abundance index and weighting factor, combined with the advantages of stratification and the
removal of some specimens to obtain biometric information, is in
keeping with the catches and knowledge on the goose barnacle
population provided by the harvesters.
Perhaps stock assessment is not the best way of assessing
S-fisheries, but it is better to use catch data to develop informative
indicators of the state of the fishery (Freire and Garcı́a-Allut,
2000). Stock population data could be useful as a complementary
method when fisheries statistics are high quality. However, to
obtain catch data is not always possible, or the quality of yield information is not adequate. In those cases, stock assessment is a
good alternative method.
The sampling methodology described in this paper proposes a
quick and simple method for stock data collection. In addition, the
data processing method yielded evaluations of the stock for specific zones and tidal levels. This procedure allows managers to estimate relative values of abundance and commercial biomass.
Handling long time series data, the absolute value could be estimated by non-linear regression, but the relative value for comparing different zones or different years is the true power of this tool.
By applying this procedure before the harvesting period, managers
could easily design rotating harvest schemes with seasonal closures
or closed areas.
Acknowledgements
L. Garcı́a, M. Miranda (technical experts at CIMA) and A. Pazo
(technical expert at the Consellerı́a do Mar) took part in the
sample processing. The suggestions of two independent referees
have improved the final version of this article.
Assessment of goose barnacle stocks in management plans
References
Barnes, M. 1996. Pedunculate cirripedes of the genus Pollicipes.
Oceanography and Marine Biology, 34: 303– 394.
Borja, A., Liria, P., Muxika, I., and Bald, J. 2006a. Relationships
between wave exposure and biomass of the goose barnacle
(Pollicipes pollicipes, Gmelin, 1790) in the Gaztelugatxe Marine
Reserve (Basque Country, northern Spain). ICES Journal of
Marine Science, 63: 626– 636.
Borja, A., Muxika, I., and Bald, J. 2006b. Protection of the goose
barnacle Pollicipes pollicipes, Gmelin, 1790 population: the
Gaztelugatxe Marine Reserve (Basque Country, northern Spain).
Scientia Marina, 70: 235– 242.
Caddy, J. F., and Defeo, O. 2003. Enhancing or restoring the
productivity of natural populations of shellfish and other marine
invertebrate resources. FAO Fisheries Technical Paper 448. Rome,
FAO. 159 pp.
Cruz, T. 1993. Growth of Pollicipes pollicipes (Gmelin, 1790)
(Cirripeda, Lepadomorpha) on the SW coast of Portugal.
Crustaceana, 65: 151 – 158.
Cruz, T., Castro, J. J., and Hawkins, S. J. 2010. Recruitment, growth
and population size structure of Pollicipes pollicipes in SW
Portugal. Journal of Experimental Marine Biology and Ecology,
392: 200 – 209.
de la Hoz, J. J., and Garcı́a, L. 1993. Datos para el estudio de la distribución y reproducción del percebe, Pollicipes cornucopiae (Leach),
en Asturias. Publicaciones Especiales del Instituto Español de
Oceonografı́a, 11: 65– 71.
Elliot, J. M. 1977. Some methods for the statistical analysis of samples
of benthic invertebrates. Freshwater Biological Association.
Scientific Publication, 25: 1 –160.
FAO. 2005. Guı́a del administrador pesquero. Medidas de ordenación
y su aplicación. Ed. by K. L. Cochrane. FAO Documento Técnico de
Pesca 424. Roma, FAO. 231 pp.
Foster, M. S., Harrold, C., and Hardin, D. D. 1991. Point vs. photo
quadrat estimates of the cover of sessile marine organisms. Journal
of Experimental Marine Biology and Ecology, 146: 193–203.
Freire, J., and Garcı́a-Allut, A. 2000. Socioeconomical and biological
causes of management failures in European artisanal fisheries:
the case of Galicia (NW Spain). Marine Policy, 24: 375– 384.
Garcı́a Allut, A. 2000. O conhecimento dos especialistas e seu papel no
desenho de novas polı́ticas pesqueiras. In Etnoconservaçao. Novos
rumos para a proteçao da natureza nos trópicos, 2nd edn, pp.
101– 123. Ed. by A. C. Diegues, and Hucitec São Paulo. 290 pp.
Haddon, M. 2011. Modelling and Quantitative Methods in Fisheries,
2nd edn. Ed. by Chapman and Hall. Boca Raton, FL. 447 pp.
Jacinto, D., Cruz, T., Silva, T., and Castro, J. J. 2010. Stalked barnacle
(Pollicipes pollicipes) harvesting in the Berlengas Nature Reserve,
Portugal: temporal variation and validation of logbook data.
ICES Journal of Marine Science, 67: 19 – 25.
1849
Jacinto, D., Cruz, T., Silva, T., and Castro, J. J. 2011. Management of
the stalked barnacle (Pollicipes pollicipes) fishery in the Berlengas
Nature Reserve (Portugal): evaluation of bag and size limit regulation measures. Scientia Marina, 75 3: 439 – 445.
Johnston, A. 1957. A comparison of the line interception, vertical
point quadrat, and loop methods as used in measuring basal area
of grassland vegetation. Canadian Journal of Plant Science, 37:
34 – 42.
Krebs, C. J. 1998. Ecological Methodology. Ed. by A. Wesley Longman,
Menlo Park, CA. 320 pp.
Lauzier, R. B. 1999. Framework for goose barnacle (Pollicpes polymerus
Sowerby, 1833) fishery in waters of the West Coast of Canada.
Fisheries and Oceans Canada. Canadian Stock Assessment
Secretariat Research. Document 99/198. 27 pp.
Molares, J. 1993. Estudio del ciclo biológico del percebe (Pollicipes
cornucopia Leach) de las costas de Galicia. Alimentaria: Revista
de tecnologı́a e higiene de los alimentos, 248 Suplement: 9 – 70.
Molares, J., and Freire, J. 2003. Development and perspectives for
community-based management of the goose barnacle (Pollicipes
pollicipes) fisheries in Galicia (NW Spain). Fisheries Research, 65:
485– 492.
Molares, J., Tilves, F., Pascual, C., and Quintana, R. 1987. Método de
determinación del tamaño y calidad del percebe (Pollicipes cornucopia, Leach). Cuadernos Marisqueros. Publicaciones Técnicas,
12: 331– 336.
Murray, S. N., Ambrose, R. F., and Dethier, M. N. 2006. Monitoring
Rocky Shores. University of California Press, Berkeley and Los
Angeles, CA. 220 pp.
Orensanz, J. M., Parma, A., Jerez, G., Barahona, N., Montecinos, M.,
and Elias, I. 2005. What are the key elements for the sustainability
of “S-fisheries”? Insights from South America. Bulletin of Marine
Science, 76: 527– 556.
Parada, J. M., and Molares, J. 2012. Artisanal exploitation of natural
clam beds: organization and management tools. In Clams
Fisheries and Aquaculture. Ed. by F. da Costa. Nova Science
Publishers Inc., New York.
Parada, J. M., Outeiral, R., Iglesias, E., and Molares, J. Consistency of
the capitulum base diameter as biometrical parameter of goose
barnacle Pollicipes pollicipes (Gmelin, 1789) (Cirripedia,
Lepadomorpha) and determination of its length at sexual maturity.
Crustaceana, in press.
Sen, S., and Nielsen, J. R. 1996. Fisheries co-management: a comparative analysis. Marine Policy, 20: 405– 418.
Shepherd, S. A., and Brown, L. D. 1993. What is an abalone stock?
implications for the role of refugia in conservation. Canadian
Journal of Fisheries and Aquatic Sciences, 50: 2001– 2007.
Thompson, S. K. 1992. Sampling. Ed. by J. Wiley and Sons, Inc. New
York, 343 pp.
www.pescadegalicia.com. (last accessed 5 March 2012).