EFAN Report 7-2000

 

 

 

Report of 2nd international workshop on hake otolith

age reading

 

Vigo, 15-19 February, 1999

 

by

 

 

Piñeiro C. (Chairperson)1 , Afonso M. H., Arego S., Bellail R., Labastie J., Loureiro I., Lucio P. ,Marecos L. ,Mc Cormick H., Moguedet Ph., Morgado C., Sainza M., Santurtún M., Trujillo V., Watson T. and Woods F.

 

 

 

1) Instituto Español de Oceanografía.

Centro Oceanográfico de Vigo

Apdo. 36 390

36280 Vigo

Spain

 

 

 

carmen.pineiro@vi.ieo.es

http://www.ieo.es

 

 

 

 

European Fish Ageing Network (EFAN)

For further information, please contact:

EFAN, Institute of Marine Research, Flødevigen Marine Research Station, N-4817 His, Norway.

Phone: (47) 37 05 90 00; Fax: (47) 37 05 90 01; Email: bente.lundin@imr.no

Office address: Flødevigvn. 49, Hisøy (Arendal), Norway.

Coordinator: Erlend Moksness, Phone (direct): (47) 37 05 90 41; E-mail: moksness@imr.no

 

1. Introduction

 

One of the main goals of the EC Study Contract No. 97/015 (DEMASSESS) is to improve the knowledge of Hake (Merluccius merluccius) growth for the northern and southern stocks.

 

The growth of European Hake from Atlantic waters has been widely studied and many researchs have been carried out to improve the knowledge on the formation and interpretation of rings in the otoliths Hickling, 1930; Bagenal, 1954, Bussy, 1966; Robles et al., 1975; Decamps and Labastie, 1978; Iglesias and Dery, 1981; Goñi, 1983; Goñi and Piñeiro, 1988; Guichet, 1988; Lucio et al. 1998a, b; Piñeiro and Hunt, 1989; Piñeiro and Pereiro, 1993; Piñeiro and Meixide, 1994; ICES CM 1983, 1984, 1986, 1996, 1997, 1999)).

 

Due to difficulties encountered in age determination, Hake has been considered by EFAN (European Fish Ageing Network) as a priority study case.

 

Taking into account the recommendations of the Workshop on Hake ageing held in Vigo in 1997, based on otoliths interpretation (Anon., 1997), it was decided to carry out an exchange of Hake otoliths from these stocks in 1998 between the countries involved in  hake stock assessment. The participants of this workshop were: IPIMAR/Portugal, IEO/Spain, AZTI/Spain, IFREMER/France, MIFRC/Ireland and CEFAS/England.

 

Conducting an age reading Workshop on Hake is considered a priority because:

 

·         Hake is of great commercial importance, which is reflected into catch value and related economy especially for Spain, France and Portugal, amongst others.

 

·         Although several International ad hoc Workshops have been devoted to otolith age reading (ICES CM 1983, 1984, 1986, 1998, 1999 and Anon., 1997) a standard criteria of Hake otolith interpretation has not yet been reached for all ages.

 

·         In recent years the ICES Working Group on the Assessment of Southern Shelf Demersal Stocks has applied numerical methods (Kimura and Chikuni, 1987) to the annual length composition of catches in order to obtain a catch at age matrix for the assessment of the Southern stock of Hake (ICES CM 1992, 1994, 1995). However, the Working Group considers the use of age length keys more reliable in order to obtain catch at age data.

 

·         Also new inexperienced staff become involved with hake age determination and need to get used to the otolith interpretation criteria.

 

In order to analyse the results of this exchange and to follow the recommendation of ICES (ICES CM 1994) and the Workshop of 1997 (Anon., 1997), a new Workshop was planned for the first quarter of 1999. The objective was to continue the work started in the 1997 Workshop  (Anon., 1997) in order to improve the agreement between readers.

 

  [1] Instituto Español de Oceanografia (IEO, Spain)

   2Instituto de Investigação das Pescas e do Mar (IPIMAR, Portugal)

  3Instituto Tecnológico, Pesquero y Alimentario (AZTI, Spain)

4Institut Français de Recherche pour l´Explotation de la Mer (IFREMER, France)

5Marine Institute Fisheries Research Centre (MIFRC, Ireland)

6Centre for Environment, Fisheries& Aquaculture Science. (CEFAS, United Kingdom) 

 

2. Objectives of the Workshop

 

·         Analysis and discussion of the results of the otoliths hake exchange in order to overcome the main problems refereed in the last workshop.

 

·         Elaborate a consensus on ageing criteria between the main laboratories involved in Northern and Southern Hake stock assessments.

 

·         Establish a protocol for Hake age determination.

 

3. Participants

 

            Afonso M. H., IPIMAR, Lisbon, Portugal

            Arego S., AZTI, Sukarrieta, Spain

            Bellail R., IFREMER, Lorient, France

            Labastie J., IFREMER, La Rochelle, France

            Loureiro I., IEO, Vigo, Spain

            Lucio P., AZTI, Sukarrieta, Spain

            Marecos L., IPIMAR, Lisbon, Portugal

            Mc Cormick H., MIFRC, Dublin, Ireland

            Moguedet Ph., IFREMER, La Rochelle, France

            Morgado C., IPIMAR, Lisbon, Portugal

            Piñeiro C., IEO, Vigo, Spain (Chairperson)

            Sainza M., IEO, Vigo, Spain

            Santurtún M., AZTI, Sukarrieta, Spain

            Trujillo V., IEO, Vigo, Spain

            Watson T., CEFAS, Lowestoff, England

            Woods F., MIFRC, Dublin, Ireland

            Alain Biseau, IFREMER, Lorient, France (Last day *)

(*)Chairman of the ICES Working Group on the Assessment of Southern Shelf Demersal Stocks.

 

 

4. Material and Methods

 

4. 1.  Material

 

The otolith collection exchange conducted during 1998 was composed of 200 Hake otolith sections from individuals ranging between 12cm and 102cm length. Two samples were available:

 

Sample 1: 100 otolith sections prepared by IEO from commercial catches sampled throughout the year and from a demersal survey conducted during the last quarter of 1997 in the Galician and Cantabrian Sea (ICES Divisions VIIIc and IXa respectively). The size ranged from 12cm to 69 cm.

 

Sample 2: 100 otolith sections prepared by IFREMER from a demersal survey conducted during the last quarter of 1997 in the Bay of Biscay and the Celtic Sea (ICES Divisions VIIIa, b and VIIf, g, h, respectively). The size ranged between 19cm and 102 cm.

 

Otoliths were mounted in black coloured resin blocks and thin sections were obtained through the nucleous level for ageing proposes. The method of otolith preparation was quite similar for both Institutes: IEO stored the otoliths in a solution of glycerine (40%) in order to enhance the rings prior to sectioning and mounted the thin sections on glass slides. IFREMER stored the otoliths dry and made several cuts for the same otolith, in order to achieve the optimum cut. In this case,  sections were not mounted on glass slides.

 

Catch date and sex information were available and also the total length in the case of IEO samples.

 

Two readings were performed during the workshop (second and third readings). Due to time constraints subsamples of the exchange sample were chosen for these readings. The size range of each subsample was maintained. The second reading consisted of 95 otolith sections, 41 from IFREMER and 54 from IEO, attempting to include well sectioned otoliths. The third reading consisted of 64 otoliths, 33 from IFREMER and 31 from IEO. Otoliths used in the second reading were excluded from the third reading in order to avoid the influence memory may have. The length frequency distribution of the samples aged is presented in Figure 1.

 

4.2 Age determination

Ageing was carried out with magnification x20, using a stereomicroscope under reflected light. Transmitted light was used occasionally. During the workshop a video camera and monitor were available for the discussion of the criteria used in otolith age interpretation. To assess whether the readers counted the same rings, it was decided to take radii measurements as per the protocol (Annex 1).

Since three sections of IFREMER samples were made of each otolith, depending on the section chosen for age determination, readers may have assigned different measurements to the same otolith. As a result it was decided only to consider the measurements of the IEO sample.

Eleven readers participated in the exchange but only ten carried out age readings in the Workshop (R6 did not read). Although all readers were experienced in reading otoliths, two participants were considered to be more experienced in ageing Hake (see table below **). Three of the readers did not participate in the previous 1997 Hake otolith workshop (see table below *).

 

During the otolith exchange the codes used for readers in the analyses were as follows:

 

READERS

 

CODE

READERS

 

CODE

 

 

 

 

Jacques Labastie (IFREMER)

R1  ** 

Maria Sainza (IEO)

R7

Lourdes Marecos (IPIMAR)

R2

Carmen Gª Piñeiro (IEO)

R8   **

Hortense Afonso (IPIMAR)

R3

Robert Bellail (IFREMER)

R9

Terry Watson (CEFAS)

R4  *

Fiona Woods (MIFRC)

R10  *

Susana Arego (AZTI)

R5

Helen Mc Cormick (MIFRC)

R11  *

Isabel Loureiro (IEO)

R6

 

 

* Not present in the first workshop in hake otoliths

** Experts in hake otoliths

 

 

The general criteria adopted for ageing each otolith are shown in the protocol (Annex 1).

These are based on the number of annual translucent rings.

In order to attain the objectives of the Workshop the following tasks were performed:

·         Analysis of the exchange results (first reading).

 

·         Discussion about individual interpretations of the otoliths, not only on those on which there was a large discrepancy amongst readers but also those on which the age assigned was the same.

 

·         Reading of the second sub-sample (95 otoliths).

 

·         Analysis of the second reading results.

 

·         Reading of the third sub-sample (64 otoliths), conducted to ascertain whether application of the discussed criteria improved agreement between readers.

 

·         Analysis of the third reading results.

 

·         Interpretation of the growth pattern for the first years of the biological cycle of the fish.

 

·        Report of the workshop.

 

In order to conduct statistical analysis it was decided to split the samples according to institute (IEO, IFREMER) due to their different characteristics (different area and preparation technique).

 

In comparison with the last workshop a more extensive analysis was performed in order to provide more details concerning individual performances. Several methods were used to analyse the results of the exchange, such as those recommended by the Workshop on Sampling Strategies for Age and Maturity (ICES CM, 1994). However, the Wilcoxon’s rank test was considered inappropriate in performing multiple paired comparisons when more than two readers are involved in ageing the same collection, which is observed in this Workshop (Zar, 1996).

 

 

4.3.  Data Analysis

 

4.3.1. Exploratory data analysis (EDA)

 

·        Determination of the modal age and of the difference between each readers’ age and the modal age. The modal age was calculated based on results from readers R1, R2, R3, R5, R7 and R8. In case of bi-modality the modal age was estimated from readers R1 and R8, the readers with most experience.

 

·         Graphical representation by reader for each sample (IEO and IFREMER), using box-whisker plots (median and interquartil range by reader). The box-whisker plots were used to summarise the observations and are useful in observing and comparing the distribution of the otolith readings by reader.

 

·         Age reading comparisons were carried out according recommendations made by ICES (ICES CM, 1994). A spreadsheet to produce the age bias plots and related tables was kindly made available by Guus Eltink (RIVO. IJmuiden. The Netherlans).

 

·         For each otolith, mean age, mode, range and standard deviation were estimated. Modal bias plots showed average age ± 2 standard deviation of each age reader and all age readers were plotted against modal age, which was considered to be the referential age. The modal age was calculated as commented before.

 

4.3.2. Computation of reproducibility measures:

 

1) Average percent age error (APE), Beamish and Fournier (1981) is an index of reading precision to compare a series of observations. The formula is as follows:

 

 

(1)

 

n  =  number of otoliths

r   =  number of readings for each otolith

xij =  the j value of age estimation for the otolith

=  average age calculated for the otolith

 

 

2) The Mean Coefficient of Variation (V)

 

 

(2)

 

sd = the standard deviation for the otolith i

 

3) The index of precision (D) (Chang, 1982):

 

 

 

(3)

 

V is described in 2)

 

These measurements are more appropriate than the conventional percent of agreement when comparing ages, since those take into account the average year class of fish.

 

 

4.3.3. Grouping readers

 

In order to determine the different groups of readers with higher agreement between them, the following statistical analyses were carried out:

 

·         Hierarchical cluster analysis using average linkage (between groups) based on squared Euclidean measure for readers without transforming the data.

 

·        Multiscaling dimension (MSD) to show the multidimensional space based on squared Euclidean measure for readers without the transformation of input data using an ordinal measure scale.

 

 

5. Results

 

5.1. First Reading

 

The results of the 200 otoliths ageing are summarised in Tables 1 and 2.

 

The Box-whisker plot for all readings from IEO sample pointed out three general groups with similar interpretations (Figure 2a):

 

1/ R1, R2, R4, R8 and R9

2/ R3, R5, R6 and R7

3/ R10 and R11

 

The Box-whisker plot for all readings from the IFREMER sample pointed out four general groups with similar interpretations but these groups are different from the previous ones (Figure 2b):

 

1/ R1, R2, R6, R7, R8 and R9

2/ R3 and R5

3/ R4

4/ R10 and R11

 

The IEO sample Modal bias plot by reader (Table 3 and Figure 3 a, b) showed that R4, R10 and R11 in general underestimated the ages. This could be due to their lack of experience in hake age determination. On the other hand, readers R3, R6 and R8 presented a tendency to overestimate the ages of the older fish. Reader R5 showed a tendency to overestimate fish of ages 2 and 3. The Modal bias plot for all readers shows some problems in age determination for age 5 and above.

 

The Modal bias plot results of the IFREMER sample (Table 4 and Figure 4 a, b) showed that the same readers (R4, R10 and R11) are still underestimating the ages. On the other hand, R9 tended to underestimate the older ages. R3 presented a tendency to overestimate fish from age 2 and above whilst R5 overestimated the younger fish (ages 2 and 3). The Modal bias plot for all readers show that ages are underestimated in relation to the modal age.

 

The APE, V and D indices are shown in Table 5. When the analysis is carried out using only the more experienced readers (R1 and R8) these indices decreased considerably. The results are different for the two samples. The IFREMER sample consists in larger individuals than the IEO sample. Thus, the IEO sample APE and V indices are higher than those for the IFREMER sample (Figure 5a-d). It should be noticed that the APE index is very sensitive to differences in younger ages.

 

The dendogram obtained from the hierarchical cluster analysis from the IEO and IFREMER samples point out the presence of two main groups, depending on the distance assumed (Figures 6a, b). In order to clarify the groups, a plot of coordinates from MSD analysis was carried out for each sample (Figures 7a, b). It can be seen from these plots that the first dimension splits also the plot into two groups, supporting the results of the hierarchical cluster analysis.

 

The two groups in each case are as follows:

 

1/ R1, R2, R3, R5, R6, R7, R8 and R9

2/ R4, R10 and R11

 

The first group consisted of readers with variable levels of experience in Hake otolith reading. The second group consisted of readers with little or no experience in reading Hake otoliths prior to this exchange, although they show extensive experience in reading otoliths of others species.

 

 

5.2 Second Reading

 

The readings of the 95 otoliths used in the second reading are presented in Tables 6 and 7.

 

The box-whisker plot for all the readings from IEO sample pointed out three groups with similar interpretations (Figure 8a): 

 

1/ R1, R7 and R8

2/ R2, R3, R4, R5, R9 and R10

3/ R11

 

The box-whisker plot for all the readings from the IFREMER sample also pointed out three groups with similar interpretations but these groups are different from the previous ones (Figure 8b):

 

1/ R1, R2, R3, and R4

2/ R5, R7, R8, R9 and R10

3/ R11

 

In both cases, R11 appears isolated from the other readers. It is to be noted that some readers changed location.

 

The IEO sample Modal bias plot by reader showed that R4, R9 and R11 in general overestimated the ages. From the first reading to the second reading they changed their criteria of interpretation (Figure 9a,b) because of their lack of experience in Hake age determination. The Modal bias plot for all readers shows improvement in ageing fish iounger than 3 years. However, problems in age determination from age 4 and above still remain.

 

The Modal bias plot results from the IFREMER sample show that the same readers are still overestimating the ages (Figure 10 a, b). R5, R7, R8 are underestimating some ages.  The Modal bias plot for all readers shows that in general ages 0 to 3 are well estimated, although problems still remain for the older ages (4 and above). Ages from 3 to 5 are overestimated in relation to the modal age (Tables 8 and 9).

 

The APE and V Indices are shown in Table 10.  The IEO reproducibility measures (APE, V and D indices) are higher than those of the IFREMER sample, but when each sample is compared with the first reading (all readers) there is a mean reduction of error of 12%.

 

The reproducibility measures obtained with the IEO otoliths collected in the 4th quarter were calculated in order to analyse whether the differences in the age attributed by readers were due to difficulties in interpreting the edge. Results obtained were similar to those obtained considering all otoliths (Table 10).

 

The dendogram from the hierarchical cluster analysis obtained in both samples, based on the distance used previously, pointed out different results (Figure 11a, b).

 

In the case of the IEO sample, there are two main groups as follows:

 

1/ R1, R2, R3, R5, R7, R8, R9 and R10

2/ R4 and R11

 

In the case of the IFREMER sample there are two groups as follows:

 

1/ R1, R2, R3, R4, R5, R7, R8, R9 and R10

2/ R11

 

Again, R11 appears isolated from the other readers. The MSD plots (Figure 12a, b) show also the

same pattern for R11, R4 and R9, which are moving around the main group.

 

It can be seen that the main group is almost stable and the less experienced readers are quite erratic/unstable.

 

The results obtained from the first and second readings are consistent. The readers with the least \experience in ageing Hake otoliths appear to have unstable criteria for their age determination. On the other hand, in general terms ages 0 to 3 are well estimated and the problems in age determination only remain for older Hake (age 4 and above).

 

 

5.3. Third Reading

 

The results of the ageing of the 64 otoliths (31 from IEO and 33 from IFREMER) used in the third reading are presented in Tables 11 and 12.

 

The Box-whisker plot for the all readings from IEO sample pointed out three main groups with similar interpretations (Figure 13 a):

 

1/ R1, R4, R7, R8 and R9,

2/ R2, R3 and R5

3/ R10 and R11

 

The Box-whisker plot for all the readings from the IFREMER sample pointed out two groups with similar interpretations but these groups are different from the previous ones (Figure 13b):

 

1/ R1, R2, R3, R5, R8, R9, R10 and R11

2/ R4 and R7

 

The IEO sample Modal bias plot by reader show that the new readers (R4, R10 and R11) tend to underestimate or overestimate ages up to 2 (Figure 14a). In general the ages older than 4 are overestimated by all readers (Figure 14b).

 

The results from the IFREMER sample show that the reader R9 is overestimating the ages up to one (Figure 15a). The Modal bias plot for all readers shows that ages from 0 to 6 are quite well estimated, although problems still remain for age 5 (Figure 15b).

 

The APE, V and D indices are presented in Table 13. The reproducibility measures obtained with the IEO sample are higher than those from the IFREMER sample. But, when each sample measures are compared to the first reading measures including all readers, there is a clear reduction of error for the IFREMER reproducibility values. As mentioned before, the difference between the results of both samples might come from the different length frequency distributions. The IEO sample consisted in more young fish than the IFREMER sample.

 

The dendograms from the hierarchical cluster analysis obtained based on the same cut distance previously used, pointed out two main groups for both samples but with different readers in each of them (Figure 16 a, b).

 

IEO sample:

1/ R1, R2, R3, R5, R7,R8, R9 and R11

2/ R4 and R10

 

IFREMER sample:

1/ R2, R3, R5, R7, R8

2/ R1, R4, R9, R10 and R11

 

The MSD plots (Figure 170 a, b) also show the general pattern found in the previous readings. A group of readers, which includes R4, R9, R10 and R11, are moving around the more experienced main group.

 

In general terms, ages 0 to 4 are well estimated and the problems in age determination remain for older Hake.

 

 

6. Discussion

 

The results obtained throughout these three readings are consistent. The readers with less experience in ageing Hake otoliths (R4, R9, R10 and R11) appear to have unstable criteria for age determination (Tables 14, 15). The differences found in the results of the third reading with respect to the second one, could be due to the low number of otoliths read (31) and the criteria used to select the third subsample. However, throughout the three readings, the non- experienced readers have changed their criteria of interpretation in relation to the consensus ageing criteria established in the previous workshop.

 

The analyses of the readings from the IFREMER samples show better agreement than those from the IEO samples. This could be due to the methodology used for the analysis in which differences in older ages cause less discrepancy than differences in younger ages.  It has to be considered that the IEO sample has a higher proportion of fish smaller than 45 cm, particularly juvenile fish, while the IFREMER sample has a greater number of old individuals.

 

As it was mentioned in the Report of the First Workshop (Anon., 1997), one of the main problems of the ageing of this species is the location of the first annual ring (Figure 18 a, b and c). It seems that in the first reading the new readers had some problems in locating that ring, the check and consequently the second ring. During the present workshop there was a remarkable improvement in identifying the same ring as the first annual ring by all readers (Figure 19). ). In general, the standardised principles for the interpretation of hake otoliths established in the first Workshop were followed by all readers.

 

Concerns were expressed from the majority of the readers regarding the poor preparation of some otolith sections, like nucleous missing, or several sections from the same otolith showing different ring structures. Respecting to the IFREMER sample, different ages could have been assigned to the same otolith depending on what section was chosen for age determination. 

 

During the Workshop, the use of various microscopes and time constraints could have been also sources of error. Different size ranges of samples in each reading could have decreased also the values of the agreement indices considered in this work. This may be particularly relevant regarding the third reading results.

 

Some readers found the interpretation of the otoliths from the Southern Hake stock more difficult, which may be related the different growth and otoliths pattern between the two stocks.

 

In terms of reproducibility measures (APE, V and D), the values for all readers in general decreased since the first reading, particularly for the IFREMER sample. V and D indices of expert readers reduced, but the APE index remained the same. It should be stressed that these measures of agreement should be interpreted with caution due to influence of sample size and of younger ages, in the case of APE.

 

Results indicate that in a near future it may be possible to use the annual ALK instead of numerical methods to estimate age composition of catches in the case of the Southern stock Hake assessment, taking into account the observed agreement between the readers involved.

 

At the moment the age structure composition of the Northern Hake stock catches are only based on IFREMER ALKs. Therefore, if an agreement on age reading is reached, the possibility of applying ALKs from AZTI in addition to IFREMER can be considered.

 

It was agreed that a comprehensive reference otolith collection representative of all ages used in the age length key should be prepared, in order to help readers and also for training purposes.

 

 

7. Conclusions

 

·         The statistical analysis shows two main groups of readers: experienced and no experienced. It is recognised the importance of experience in Hake ageing.

 

·         The agreement between readers for ages 0 to 4 improved comparatively to the first Hake Workshop. This can be a result of the adoption of the ageing criteria established in that Workshop.

 

·         There was a high level of variability between readers for ages 5 and older.

 

·         A high agreement in locating the first annual ring between readers was achieved.

 

·         The less experienced readers showed improvement in their age interpretation criteria.

 

·         Despite the improvement of all the readers, the results highlight the difficulties in interpreting the age of Hake otoliths and justify the need for periodical exchange exercises.

 

 

8. Recommendations

 

·         Results indicate that may be possible to use the annual ALK, instead of numerical methods, to estimate the age composition of catches of the Southern Stock of Hake.

 

·         In order to use the same ALK for Northern stock assessment, the age estimation criteria should be the same for AZTI and IFREMER.

 

·         Informal exchanges should be carried out between Institutes on a regular basis, especially for those who supply ALKs to the assessment Working Group.

 

·         Images of sectioned otoliths from the exchange collection will be digitised and interpreted by the more experienced readers. This will act as both a reference collection and an age determination guide.

 

·         Another otolith exchange and Workshop should be convened in the next two years.

 

·         Statistical tests on age data of each Institute should be carried out before they are combined for ALKs.

 

·         It is desirable that at least two people per Institute should be involved in Hake age determination. Also some more countries should take part in these studies.

 

·         Otoliths used in an exchange should be prepared following the same methodology and suitable for transportation.

 

·         It is necessary to validate the age estimation of this species by conducting more studies on the life history events of the fish, in addition to exploring alternative validation techniques (tagging, microchemistry, etc.). These recommendations require a dedicated project on these matters.

 

 

9. Contributions to the Workshop

 

Additional information was presented during the Workshop, in order to improve the age determination of this species:

 

-          Quarterly length composition (in percentage) of juvenile Hake (<30 cm) catches, including discards by bottom trawlers in the Bay of Biscay, for the period 1988-1997, from AZTI. (Values for the 3rd quarter of 1988 were estimated from the 2nd and 4th quarters) (Annex II).

 

-          Preliminary results on daily growth of juvenile hake, from IPIMAR (Annex III).

 

-          A training guide to introduce the readers to the age reading criteria established for hake with digitised images of otolith sections from ages 0 to 4 (Annex IV).

 

 

10. References

 

Anon. 1997. Report of the Workshop on Hake otoliths age readings. EC Study Contract no. 95/038 “Biological studies of Demersal Fish (BIOSDEF)12pp.

Bagenal, T.B. 1954. Growth rate of the hake (Merluccius merluccius L.) in the Clyde and other Scottish areas. J. Mar. Biol. Ass, U.K., 33(1): 69-95.

Beamish, R. J. and Fournier, D. A. 1981. A method for  comparing the precision of a set of age determinations. Can. J. Fish. Aquat. Sci. 38 : 982-983.

Busy. M. M., 1966. La croisance du merlu dans le Golfe de Gascogne. ICES, C.M. 1966/G:17.

Chang, W. Y. B., 1982. A statistical method for evaluating the reproducibility of age determination. Can. J. Fish Aquat. Sci. 39 : 1208-1210.

Decamps, P. and J. Labastie, 1978. Note sur la lecture et l'interpretation des otolithes du merlu. ICES, C.M. 1978/G:41.

Eltink, G., 1994. Comparison of otolith readings. Working document for the Workshop on sampling strategies for age and maturity, 1994 at ICES Copenhagen.

Goñi, R., 1983. Growth studies of European hake (Merluccius merluccius L.) from the Northwest African shelf. Demersal Fish Committee C.M. 1983/G10.

Goñi, R. and C. Piñeiro, 1988. Study of the growth pattern of European hake Merluccius merluccius L.) from the Southern stock: ICES Divisions VIIIc and IXa. ICES, C.M. 1988/G:18.

Guichet, R. 1988. Etude de la croissance du merlu europeen (Merluccius merluccius L.) au cours de ses premières années. Analyse per NORMSEP des distributions en taille observées trimestriellement en mer de 1980-87. Working Paper in ICES, C.M. 1988/WP assess:24.

Hickling, C.F., 1930. The natural history of the hake. Part III. Seasonal changes in the condition of hake. Fish. Invest., Ser. II, XII (1).

ICES CM, 1983. Confrontation Technique sur les Methodologies Utilisées pour la Lecture des Otoliths de Merlu. La Rochelle, France, 7-13 June, 1983. Appendix of Report of the Working Group on Assessment of Hake Stocks. ICES, C.M. 1983/Assess:2.

ICES CM, 1984. Report of the ad hoc Hake Otolith Working Group, La Rochelle, France, 28 May - 1 June 1984. ICES, C.M. 1984/G:74.

ICES CM, 1986. Report of the ad hoc Study Group on Hake, Lowestoft,2-6 June 1986. ICES, C.M. 1986/G:93.

ICES CM, 1992. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks, Copenhagen, 8-17 September 1992. ICES, C.M. 1993/Assess:3.

ICES CM, 1994. Report of the Workshop on Sampling Strategies for Age and Maturity. ICES, C.M. 1994/D: 1, 67 pp.

ICES CM, 1994. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM. 1994/Assess 3: 447 pp.

ICES CM, 1995. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM. 1995/Assess 6: 598 pp.

ICES CM, 1996. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM 1996/Assess 5: 493 pp.

ICES CM, 1997. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM 1997/Assess 5: 640 pp.

ICES CM, 1998. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM 1998/Assess 4: 664 pp.

Iglesias, S. and L. Dery, 1981. Age and growth studies of hake (Merluccius merluccius L.) from ICES divisions VIIIc and IXa. ICES, C.M. 1981/G:38.

Kimura, D. K. and S. Chikuni. 1987. Mistures of the empirical distributions: an iterative application of the age-length key. Biometrics. 13: 23-35.

Lucio, P., Murua, H. and Santurtún, M., 1998a. Growth and reproduction of hake (Merluccius merluccius) in the Bay of Biscay during the period 1996-1997. Crecimiento y reproducción de la merluza (Merluccius merluccius) del Golfo de Vizcaya en el período 1996-1997. Sexto Coloquio Internacional sobre Oceanografía del Golfo de Vizcaya. Donostia-San Sebastián. Abril 1998. (in press in OZEANOGRAFIKA)

Lucio, P., Santurtún, M., and Murua, H., 1998b. Growth and reproduction of hake (Merluccius merluccius) in the Bay of Biscay during 1996-1997. I.C.E.S. C.M. 1998/CC:20. 24p+12.

Piñeiro, C. and J. J. Hunt, 1989. Comparative study on growth of european hake (Merluccius merluccius L.) from Southern stock using whole and sectioned otoliths, and length frequency distributions. ICES, C.M. 1989/G:37.

Piñeiro, C. and M. Meixide, 1994. Analysis of results obtained in the Spanish-French meeting about otolith age reading of european hake. Working Paper in ICES, C.M. 1994.

Piñeiro, C.G. and J.A. Pereiro, 1993. Study on juvenile growth pattern of European hake (Merluccius merluccius L.) using whole otoliths and length frequency distributions from commercial catches and fish surveys. ICES C.M./G:12.

Robles, R, Pereiro, J. A., Iglesias, S. et Pereiro, F. J., 1975. Étude sur la croissance du merlu europeen, Merluccius merluccius L. du NW de L’Espagne. C.M. 1975/ G: 8 Comité des ICES.

Zar, J. H. 1996. Biostatistical analysis. Prentice Hall, New Jersey, 3rd edition, 662pp.

 

The full EFAN Report is available for download using this link EFAN Report 7/2000.doc (the file is 12MB)