Skip Navigation


ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on November 6, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(9):1785-1790; doi:10.1093/icesjms/fsm163
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
64/9/1785    most recent
fsm163v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Martins, M. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Martins, M. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2007 International Council for the Exploration of the Sea. Published by Oxford Journals. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Growth variability in Atlantic mackerel (Scomber scombrus) and Spanish mackerel (Scomber japonicus) off Portugal

Maria Manuel Martins

Instituto de Investigação Agrária e das Pescas (INIAP), Av. Brasília, 1449-006 Lisbon, Portugal

tel: +351 21 3027000; fax: +351 21 3015948; e-mail: mane{at}ipimar.pt

Martins, M. M. 2007. Growth variability in Atlantic mackerel (Scomber scombrus) and Spanish mackerel (Scomber japonicus) off Portugal. – ICES Journal of Marine Science, 64.

Year-class effects may be very important when analysing fish growth. This work aims to compare growth in weight and in length between year classes of Atlantic and Spanish mackerel found off the Portuguese coast. The analysis included Atlantic mackerel data from year classes 1986–1993 and Spanish mackerel data from year classes 1988–1995. The analyses indicated significant differences between year classes in terms of mean weight-at-length and mean length-at-age. Atlantic mackerel year classes also showed significant differences in mean weight-at-length between males and females, and Spanish mackerel year classes showed significant differences between sexes for mean length-at-age. No significant effects were found for the two-way interaction between year class and sex. Growth variability may be linked with variability in environmental factors as well as density-dependence.

Keywords: age, Atlantic mackerel, growth, length–weight, Portuguese coast, Spanish mackerel, year classes

Received 19 November 2006; accepted 4 October 2007; advance access publication 6 November 2007.


   Introduction
Top
 Introduction
Material and methods
 Results
 Discussion
 References
 
Atlantic mackerel (Scomber scombrus) and Spanish mackerel (Scomber japonicus) have a wide distribution throughout the Atlantic Ocean, though the former is more abundant at more northern latitudes than the latter (Whitehead et al., 1984). The same pattern is observed in Portuguese waters, where Atlantic mackerel have a more northerly distribution than Spanish mackerel (Martins and Cardador, 1996). In Portuguese waters, both species are targeted by fisheries, but Atlantic mackerel is less abundant than Spanish mackerel. Average annual catches of the two species between 1986 and 2003 were 3200 and 7200 t, respectively.

The growth of Atlantic and Spanish mackerel off Portugal has been studied. Gordo et al. (1982) and Martins and Gordo (1984) compared growth rates of the two species from different locations off Portugal, and Martins et al. (1983) compared Spanish mackerel mean lengths-at-age between Portugal and several other locations throughout the world. Martins (1998a) made a comparative study of the first ring of the otolith of Atlantic mackerel using samples from the Portuguese coast and other areas to the north.

Most growth studies on the two species have analysed data by year or quarter. However, growth models try to represent the growth of individual fish, so are meaningful only when fitted to a year class, not to an aggregation of several year classes at a given time. The year-class effect may be important when comparing growth curves (Dawson, 1986), and it cannot be detected when analysing aggregated data.

The objective of this work was to compare year classes with respect to growth for both species of mackerel off Portugal. Two types of analysis were carried out for each: (i) mean weight-at-length comparison by year class for both sexes, and (ii) analysis of mean length-at-age by year class for both sexes. The overall objective of the study was to compare growth in terms of length and weight off the coast of Portugal between year classes of both species, and to analyse the factors that influence the growth patterns of those year classes.


   Material and methods
Top
 Introduction
 Material and methods
Results
 Discussion
 References
 
Monthly samples of both species were obtained from commercial catches and scientific surveys in ICES Division IXa, from 1986 to 2003. Atlantic mackerel belonging to the year classes from 1986 to 1993 (3682 fish for weight/length and 4741 for age/length) were included. Total length ranged from 17.8 to 46.5 cm, and fish total weight from 39.2 to 854 g. The Spanish mackerel analysed (3259 fish for weight/length and 5243 for age/length) belonged to year classes from 1988 to 1995. Their total lengths ranged from 13.0 to 41.9 cm and their total weight from 23.4 to 723.9 g. The number of fish per year class is listed in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Numbers of Atlantic mackerel and Spanish mackerel from the 1986–1995 year classes included in the ANCOVA analyses.

 
Each fish was measured to the nearest millimetre and weighed to the nearest 0.1 g. Sex was determined and otoliths removed for age reading. For Atlantic mackerel, age within the range 0–15+ years was determined according to the method described in ICES (1995). The same criteria were used for Spanish mackerel otoliths, because they have a similar structure, though they differ in the shape of the sagittae (the otoliths). As for Atlantic mackerel, growth rings can be observed on the external face, and opaque areas of rapid growth and hyaline or transparent rings are formed annually. The period of fast growth for Spanish mackerel is summer, and winter and spring are considered seasons of slower growth (Rodriguez-Roda, 1982; Martins et al., 1983; Aguayo and Steffens, 1986; Forciniti and Perrotta, 1988; Perrotta, 1992).

The relationship between weight and length was evaluated by analysis of covariance (ANCOVA). The natural logarithms of the data (ln) were used as follows: ln (W) = a + b ln (L) + sex + year class + error, where W is the total weight (g) and L the total length (cm). In this case, ln (weight) was the dependent variable, and ln (length), year class, and sex the explanatory variables.

Growth in length was examined by analyses of covariance on length-at-age data by year class, with fish length as dependent variable, and year class, sex, and age as explanatory variables. Analyses of weights and lengths included ages from 0 to 10 years for Atlantic mackerel, and from 0 to 8 years for Spanish mackerel. Statistical analyses were performed using the software package StatisticaTM (Statsoft, 2001).

As the analyses of covariance revealed year-class effects, Newman–Keuls tests (Zar, 1996) were applied to the year-class means of the ln (weights-at-length) and mean lengths-at-age, to trace the source of the year-class effects by identifying year classes with similar growth properties. The relationship between early growth and year-class strength was examined by Pearson correlation analysis. For each species, mean lengths at each of ages 0–3 were correlated with stock abundance of the year class (total numbers-at-ages 0 and 1). The abundances of year classes were taken from stock assessments (ICES, 2004, for Atlantic mackerel; Martins, 1998b, for Spanish mackerel). Mean length-at-age was calculated from monthly samples, weighted by the sample sizes.


   Results
Top
 Introduction
 Material and methods
 Results
Discussion
 References
 
ANCOVA was used for statistical analysis, because normal probability plots of the residuals of total length, ln (total length), and ln (total weight) revealed no serious violation of the normality assumption. The results for weight-at-length show that mean weight varies significantly between year classes for both Atlantic and Spanish mackerel (Table 2). The analysis of covariance for length-at-age also shows that mean length-at-age varies significantly between year classes (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 2. Weight–length analysis of covariance for Atlantic mackerel year classes 1986–1993 and Spanish mackerel year classes 1988–1995.

 


View this table:
[in this window]
[in a new window]

  		Table 3. Age–length analysis of covariance for Atlantic mackerel year classes 1986–1993 and Spanish mackerel year classes 1988–1995.

 
Using the Newman–Keuls test on mean ln (weight), the Atlantic mackerel year classes 1991, 1992, and 1993 formed one group with no significant differences between them. Spanish mackerel of the 1988, 1989, and 1992 year classes also had no significant differences between them. The Newman–Keuls test on mean lengths revealed that the Atlantic mackerel year classes 1986, 1987, 1989, 1991, and 1993 formed one group with no significant differences between them, and that the Spanish mackerel year classes 1988, 1989, 1990, 1994, and 1995 were not significantly different.

Differences in growth in length between Atlantic mackerel year classes were most prominent in the first years of life (Figure 1), where the difference is >2 cm. The year classes with the lowest length-at-ages 1 and 2 were the 1987, 1988, 1992, and 1993 year classes. The 1989 year class seems to have grown slowly also at ages 0 and 3. At greater age, there is virtually no difference in mean length-at-age between the year classes.


Figure 1
View larger version (16K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Atlantic mackerel male and female mean length-at-age for the year classes 1986–1993.

 
For Spanish mackerel, the differences in mean length-at-age between year classes were also most prominent at the youngest ages (Figure 2). Fish aged 0, 1, and 2 belonging to the year classes born in 1992, 1993, 1994, and 1995 had the lowest mean length-at-age, and age 1 mackerel from the 1991 year class also had a low mean length. Both sexes grew more slowly after the third year of life, the age at which 50% first maturity is attained (Martins et al., 1983; Martins, 1996).


Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 2. Spanish mackerel male and female mean length-at-age for the year classes 1988–1995.

 
The analysis of covariance showed significant differences in ln (mean weight-at-length) between Atlantic mackerel males and females, but not for mean length-at-age. For Spanish mackerel, there was a significant difference between sexes for mean length-at-age, but not for ln (mean weight-at-length) (Tables 2 and 3). Figure 3 shows the two-way interaction between year class and sex (mean and s.d. of ln mean weights-at-length) for each year class for Atlantic mackerel. The differences between sexes are not significant when the two factors interact, as also indicated in Table 2. Moreover, there is no clear trend in ln (mean weight) over the period analysed. Single weight–length relationships for males and females of each year class showed that, in general, the differences are larger for the youngest mackerel, <30 cm total length.


Figure 3
View larger version (11K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 3. Atlantic mackerel in weight by year class, 1986–1993.

 
Female Spanish mackerel year classes usually have a higher mean length-at-age than males. The mean length decreased during the period analysed for both sexes. Differences between sexes are not significant when the two factors interact (Figure 4). Correlations between mean length at each of ages 0–3 and the abundance of year classes 0 and 1 (estimated by analytical assessment as numbers-at-ages 0 and 1) for Atlantic mackerel are always negative, but never statistically significant (Table 4).


Figure 4
View larger version (8K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 4. Spanish mackerel mean length by year class and sex.

 


View this table:
[in this window]
[in a new window]

  		Table 4. Results of the correlation of mean length-at-age against stock size (numbers aged 0 or 1) of Atlantic mackerel year classes, 1986–1993.

 
Similar analyses were carried out on the data for Spanish mackerel. Mean lengths-at-age for each of ages 0–2 (mean length-at-age 3 is positively correlated either with numbers at age 0 or 1), and the magnitude of their year class at ages 0 and 1 (Table 5). Although the correlations were always negative, they were not statistically significant, except for age 1 numbers and age 2 mean lengths (p < 0.05).


View this table:
[in this window]
[in a new window]

  		Table 5. Results of the correlation of mean lengths-at-age against stock size (numbers aged ages 0 or 1) of Spanish mackerel year classes, 1988–1995.

 

   Discussion
Top
 Introduction
 Material and methods
 Results
 Discussion
References
 
Results show that there are significant differences between year classes with respect to mean weight-at-length for both species. The growth and maturation of all fish is highly dependent on the environment (food availability, temperature, etc.). Temperature influences the population dynamics of species through growth (Brander, 1995; Brander et al., 2003). It is likely that differences can be detected between year classes because the individuals were spawned under different environmental conditions, e.g. temperature. For the period 1980–2000 west of Portugal, sea surface temperatures (SSTs) displayed a positive anomaly when compared with the period 1960–1980 (Borges et al., 2004). During the period 1987–1990, temperature was above average, and from 1991 to 1994, it was below average. Atlantic mackerel and Spanish mackerel are considered to be a boreal and a tropical species, respectively, so should therefore react differently under the same environmental conditions.

The other analysis investigated the differences between growth in length of the year classes of both species, and revealed significant differences in mean length-at-age between year classes, for both. Such differences have previously been reported by Dawson (1991), who detected significant differences in first year growth (L1) of Atlantic mackerel otoliths belonging to different year classes.

Analysing year class length-at-age from both species and SST for the period 1960–2000 off western Portugal, we see that the response by Spanish and Atlantic mackerel was different in years where temperature was high. The response of Spanish mackerel (a tropical species; Borges et al., 2004) was positive (year classes from 1988 to 1990, and 1995). Atlantic mackerel year classes born during 1986, 1991, and 1993 also showed a positive response, but only when the temperature was below average (it is a boreal species; Borges et al., 2004).

The density-dependence phenomenon, often cited in the literature as contributing to variability in growth in length or weight (Hamre, 1978; Overholtz, 1989), was also considered. Overholtz (1989) obtained high correlation coefficients between mean weights-at-age of mackerel 1 and 2 years old and year-class size-at-age, although the relationships were not significant because of the small number of samples analysed. Additionally, Dawson (1991) found a negative correlation between mean L1 length and total stock size, in numbers, of North Sea mackerel, suggesting a density-dependent mechanism. The negative correlation between ages 0 and 1 numbers and lengths-at-age for young Atlantic and Spanish mackerel found here agrees with literature statements, although just one of the correlations was individually significant. This finding is compatible with the hypothesis of density-dependent growth of the two species. The low mean length-at-age of the young Atlantic mackerel shown in Figure 1 may be partly caused by density-dependence. Year classes 1987, 1992, and 1993 were the largest ones in the record, and the corresponding mean lengths were the smallest. For Spanish mackerel too, the trend of mean length-at-age shown in Figure 2 could be partly caused by density-dependence: year classes 1992, 1993, and 1994 were the largest in the record, and young fish of these year classes had the smallest length-at-age. Moreover, temperature during those years was below average and could have had a negative influence on growth. The low length-at-age of the 1995 year class could also perhaps be attributable to an influence of the preceding year class, which was the largest in the record.

Atlantic and Spanish mackerel are both oceanic species with aggregating and migratory tendencies. Differences found between cohort growths mainly at young ages may be influenced by local environmental conditions, which diminish when the fish mature and begin to migrate over the great distances they are known to cover. Studies of these issues have been carried out by a number of authors (e.g. Kramer, 1969; Walsh and Martin, 1986; Cousseau et al., 1987; Walsh et al., 1995). Reid (2001) describes mackerel distribution and migration linked to the oceanographic environment off the European shelf edge. Additionally, the main conclusions of the EU project "Spatial Pattern of Migration and Recruitment of North East Atlantic Mackerel" clearly show mixing and migration of mackerel from off the Iberian Peninsula to as far as Norwegian waters, while juveniles stay closer to the areas where they were tagged (Uriarte et al., 2001).

Reid (2001), referring to mackerel analysed during the SEFOS (Shelf Edge Fisheries and Oceanography Studies) project, stated that the spatial and temporal distribution of spawning appeared to have changed slightly over the past 20 years. An analysis of the distribution of eggs and adults during the spawning season indicated that these changes were influenced directly by oceanographic conditions, which in turn affected the recruitment success recently (Walsh et al., 1996).

The distribution and migration of Spanish mackerel have been studied in the Southeast and Southwest Atlantic (Cousseau et al., 1987; Ostapenko, 1988), but not off the Iberian Peninsula and adjacent areas. However, mackerel in their prespawning and early post-spawning stages are commonly sampled there, and fish in spawning condition are seldom found in samples collected during the spawning season off Portugal (Martins et al., 1983; Martins, 1996). It is therefore likely that the Spanish mackerel spawning grounds are south of the Portuguese coast, and if that is the case, year class strength would be influenced by the environmental conditions there.

In conclusion, there are differences in growth between year classes, mostly at young ages, of both species of mackerel. There is also some evidence that these differences may be related to environmental variations and year-class strength. There are also some differences in growth between sexes, but the reasons for those remain unclear.


   Acknowledgements
 
I thank my colleagues for support in writing this paper, especially Alberto Murta and João Pereira for their useful, highly valid comments. Also, I thank referees D. Skagen and L. Gordo for their comments, which helped improve the manuscript greatly. Some of the samples analysed were obtained through the PNAB/Data Collection Programme.


   References
Top
 Introduction
 Material and methods
 Results
 Discussion
 References
 

    Aguayo H. M., Steffens A. H. Edad y crecimiento de Scomber japonicus del norte de Chile. Investigación Pesquera, Santiago (1986) 33:61–76.

    Borges M. F., Hill L., Mendes H., Santos A. M., Salgado M. Regime shift effects on fish community metrics measured in the West Portugal upwelling area. ICES Document CM 2004/M: 11. (2004) 13.

    Brander K. M. The effect of temperature on growth of Atlantic cod (Gadus morhua L.). ICES Journal of Marine Science (1995) 52:1–10.[Abstract/Free Full Text]

    Brander K., Borges M. F., Erzini K., Henderson G., Mackenzie B. R., Mendes H., Ribeiro J., et al. Changes in fish distribution in the eastern North Atlantic: are we seeing a coherent response to changing temperature? ICES Marine Science Symposia (2003) 219:261–270.

    Cousseau M. B., Angelescou V., Perrota R. G. Algunas caracteristicas de la estructura y comportamiento migratorio de los cardumenes de caballa (Scomber japonicus marplatensis) en la plataforma bonaerense (Mar Argentino), periodo 1965–1984. Revista de Investigación y Desarrollo Pesquero (INIDEP) (1987) 7:21–42.

    Dawson W. A. Changes in western mackerel (Scomber scombrus) spawning stock composition during the spawning season. Journal of the Marine Biological Association of the UK (1986) 66:367–383.

    Dawson W. A. Otolith measurement as a method of identifying factors affecting first-year growth and stock separation of mackerel (Scomber scombrus L). Journal du Conseil International pour l’Exploration de la Mer (1991) 47:303–317.

    Forciniti L., Perrotta R. G. Sobre la edad y el crecimiento de la caballa (Scomber japonicus) del area marplatense. Revista de Investigación y Desarrollo Pesquero (INIDEP) (1988) 8:19–32.

    Gordo L. S., Martins M. M. B., Jorge I. M. Preliminary study on the age and growth of mackerel (Scomber scombrus L.) in ICES Sub-area IX. ICES Document CM 1982/H: 16. (1982) 13.

    Hamre J. The effect of recent changes in the North Sea mackerel fishery on stock and yield. Rapports et Procès-verbaux des Réunions du Conseil International pour l’Exploration de la Mer (1978) 172:197–210.

    ICES. Report of the workshop on mackerel otolith reading. ICES Document CM 1995/H: 1. (1995) 45.

    ICES. Report of the Working Group on the Assessment of Mackerel, Horse Mackerel, Sardine and Anchovy. ICES Document CM 2004/ACFM: 08. (2004) 493.

    Kramer D. Synopsis of the biological data on the Pacific mackerel, Scomber japonicus Houttuyn (Northeast Pacific). FAO Species Synopsis (1969) 40.:18.

    Martins M. M. New biological data on growth and maturity of Spanish mackerel (Scomber japonicus) off the Portuguese coast (ICES Division IX a). ICES Document CM 1996/H: 23. (1996) 17.

    Martins M. M. S. scombrus L. from ICES Div. IX a (Portugal): analysis on otolith first ring and an infestation study by Anisakis simplex (Nematoda: Ascaridae). ICES Document CM 1998/CC: 17. (1998) a:16.

    Martins M. M. As Populações do Género Scomber: Sarda (S. scombrus L. 1758) e Cavala (S. japonicus, H. 1782). Biologia e estado de conservação destes recursos nas áreas de distribuição do Atlântico Nordeste. Dissertação original apresentada para provas de Investigadora Auxiliar. Área Científica de Dinâmica de Populações Haliêuticas do Instituto de Investigação das Pescas e do Mar. Lisboa, Julho de 1998. (1998) b:146.

    Martins M. M., Cardador F. Abundance and distribution pattern of Spanish mackerel (S. japonicus) and mackerel (Scomber scombrus L.) in the Portuguese continental waters (ICES Div: IXa). ICES Document CM 1996/H: 24. (1996) 21.

    Martins M. M., Gordo L. S. On the comparison of Spanish mackerel (Scomber japonicus Houttuyn 1782) from Gorringe Bank and Peniche (Portuguese coast). ICES Document CM 1984/H: 50. (1984) 7.

    Martins M. M., Jorge I. M., Gordo L. S. On the maturity, morphological characteristics and growth of Scomber japonicus Houttuyn (1782) of west continental coast of Portugal. ICES Document CM 1983/H: 39. (1983) 9.

    Ostapenko A. T. Age, croissance et caracteristiques du maquereau espagnol (Scomber japonicus Houtt.) de l’Atlantique Sud-est. Collection of Scientific Papers International Commission for the Southeast Atlantic Fisheries (1988) 15:161–174.

    Overholtz W. J. Density-dependent growth in the Northwest Atlantic stock of Atlantic mackerel (Scomber scombrus). Journal of Northwest Atlantic Fisheries Science (1989) 9:115–121.

    Perrotta R. G. Growth of mackerel (Scomber japonicus Houttuyn, 1782) from Buenos Aires—north Patagonian region (Argentine Sea). Scientia Marina (1992) 56:7–16.

    Reid D. G. SEFOS—shelf edge fisheries and oceanography studies: an overview. Fisheries Research (2001) 50:1–15.[CrossRef][Web of Science]

    Rodrigues-Roda J. Biologia de la caballa (o estornino), Scomber (Pneumatoforus) japonicus Houttuyn (1782) del golfo de Cádiz. Investigaciones Pesqueras de Cadiz (1982) 46:143–149.

    Statsoft. STATISTICA System Reference. (2001) USA: Statsoft, Inc. 1098.

    Uriarte A., Alvarez P., Iversen S., Molloy J., Villamor B., Martins M. M., Myklevoll S. Spatial pattern of migration and recruitment of North East Atlantic mackerel. ICES Document CM 2001/O: 17. (2001) 40.

    Walsh M., Martin J. H. A. Recent changes in distribution and migrations of the western mackerel stock in relation to hydrographic changes. ICES Document CM 1986/H: 17. (1986) 7.

    Walsh M., Reid D., Turrel W. R. Understanding mackerel migration off Scotland: tracking with echosounders and commercial data, and including environmental correlates and behaviour. ICES Journal of Marine Science (1995) 52:925–939.[Abstract/Free Full Text]

    Walsh M., Skogen M., Reid D. G., Svendsen E. The relationship between the location of mackerel spawning, larval drift and recruit distributions: a modelling study. ICES Document CM 1996/S: 33 (1996).

    Whitehead P. J. P., Bauchot M. L., Hureau J. C., Nielsen J., Tortonese E. Scombridae. In: Fishes of the North-eastern Atlantic and the Mediterranean (1984) Paris: UNESCO. 981–997.

    Zar J. H. Biostatistical Analysis. (1996) Prentice Hall International Editions. 662.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
64/9/1785    most recent
fsm163v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Martins, M. M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Martins, M. M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?