Elsevier

Fisheries Research

Volume 78, Issue 1, April 2006, Pages 96-106
Fisheries Research

The English Channel stock of Sepia officinalis: Modelling variability in abundance and impact of the fishery

https://doi.org/10.1016/j.fishres.2005.12.004Get rights and content

Abstract

The English Channel cuttlefish fishery provides one of the highest cephalopod yields in the north-east Atlantic, with catches by many interacting fishing gears. The aim of this study is to show that population modelling can be developed for this stock and that model estimates are useful to address practical issues such as diagnostics of fishing pressure and interactions between fishing fleets. Abundance indices were calculated from catch and effort data using a General Linear Model and stock size was assessed with analytical methods based on biological data collected since 1996. Cohort analysis was carried out on six cuttlefish cohorts on a monthly basis using catch-at-age data and the results were integrated in a Thomson and Bell model to simulate production and biomass of the stock. GLM indices indicated a strong seasonal pattern of abundance with a peak of adult abundance in spring and of juveniles in autumn. Cohort analysis estimates show that recruitment can vary by a factor of 2 between years. Estimated fishing mortalities indicate rather stable exploitation patterns. Diagnostics show that each studied cohort was generally fully exploited, and do not reveal evidence of recruitment over-fishing. Simulations of fishing fleet interactions for the resource show that otter trawls play a dominant role in cohort exploitation whereas artisanal trap net fisheries are described as “dependent” on other gears. In spite of biological differences between teleost fishes and cephalopods, classical age-based models can be adapted to cuttlefish to provide fishery management advice. Although marked over-fishing was not observed, the studied cohorts were subjected to high catch rates and there is little scope for increased exploitation. Interactions between the fishing fleets involved underline the point that management in the coastal zone alone is not sufficient to ensure sustainable exploitation in a migrating species.

Introduction

The depletion of finfish stocks in the north-east Atlantic makes desirable the assessment of alternative resources such as cephalopods, with the aim of developing rational management procedures. Indeed, cephalopod stocks are, as yet, not generally subjected to management measures in European waters. Some EC countries impose minimum landing sizes, and European funding for national fishery data collection programs (Council Resolution 1543/2000) includes some provision for collection of data on cephalopod fisheries, notably in Spain and Portugal.

Modelling and understanding cuttlefish exploitation is a challenge for a series of reasons: life-cycle and population dynamic are different from those of finfish (Caddy, 1983, Boyle and Boletzky, 1996) but also of squid (Boyle, 1983), fishing is carried out by a range of gears and interacting fishing fleets, and cuttlefish distribution and abundance are influenced by environmental variation (Boletzky, 1983, Sobrino et al., 2002, Wang et al., 2003).

In terms of yields, cuttlefish is currently the most important cephalopod group taken in the north-east Atlantic, and the main fishing ground for this resource is the English Channel (International Council for the Exploration of the Sea-ICES-divisions VIId and VIIe). On average, annual landings reached 10 500 tonnes in 1993–2002 (Anon., 2003), which represents 60% of the total production of cuttlefish from ICES waters. The resource shows marked interannual and seasonal variability in catches, a phenomenon common to most fisheries of short-lived species, reflecting changes in local abundance (see Wang et al., 2003). Cuttlefish is mainly fished by French and English multispecies demersal trawlers and, seasonally, by coastal traps. Catches by all gears peak in autumn and in spring (Denis and Robin, 2001, Royer, 2002).

In the English Channel, Sepia officinalis Linnaeus, 1758 is the only cuttlefish caught by commercial trawlers and the bulk of the population has a 2-year cycle (Boucaud-Camou et al., 1991). The species spawns in spring on both north (English) and south (French) coasts (Boucaud-Camou and Boismery, 1991, Dunn, 1999). The main period of recruitment to the fishery is October–November, when young of the year migrate offshore to wintering grounds. A second group of recruits is observed in March–April when the stock migrates to inshore areas. These recruits are considered to be animals born late in the hatching season, which experienced slower juvenile growth than the first group (Medhioub, 1986, Royer, 2002). The stock is thus composed of two overlapping annual cohorts (including two sub-groups of recruits in each cohort).

Boundaries for the stock (ICES divisions VIId and VIIe) were determined from several types of observations. Biologically, the life-cycle takes place almost entirely within the English Channel and this area represents the northern limit for reproduction of the species. Moreover, the migratory pattern described by Boucaud-Camou and Boismery (1991) is coherent with the assumption of an English Channel stock unit. From a fishery point of view, cuttlefish concentrate within the English Channel and Catch Per Unit Effort (CPUE) is low in adjacent waters, in the southern North Sea and in the Celtic Sea (Denis, 2000). Therefore, the English Channel is considered as a potential management unit, as for loliginid squid stocks (Royer et al., 2002).

Variability in abundance of stocks of marine animals can either be described using indices, normally derived from catch and fishing effort, or using estimates of absolute stock size. Catch and effort data have been extensively used to compute abundance indices that describe stock trends (Chadwick and O’Boyle, 1990, Goni et al., 1999, Gordoa et al., 2000, Punt et al., 2000). However, catch rates (Catch Per Unit Effort) derived from fishery statistics require standardization to take into account variation in fishing power. Indeed, the relationship between raw CPUE and abundance can be influenced by several factors, including heterogeneous spatial distribution of the resource, and differences in fishing fleet composition and in fishing equipment (Hilborn and Walters, 1992). General Linear Modelling techniques (GLM) have nowadays superseded previous CPUE weighting techniques (Hilborn and Walters, 1992).

In cephalopods, the most commonly used stock assessment method is the depletion method (see Pierce and Guerra, 1994, Young et al., 2004). However, the underlying assumptions of the abundance model are restrictive, for example it is usually assumed that catchability of the animals is constant over the whole studied period. This hypothesis is likely to be inappropriate in the case of English Channel cuttlefish. This stock undergoes marked seasonal migratory movements (Boucaud-Camou and Boismery, 1991) and catchability is likely to differ according to location (i.e. in offshore or inshore fishing grounds). Furthermore, migrations are related to ontogenetic development, which is accompanied by a change in cuttlefish behaviour (Hanlon and Messenger, 1996). Analytical assessment methods do not require such strong assumptions but assume that total catch can be estimated by age-class. They are less often used for cephalopod stocks since, in these populations, life-span is short, growth is fast and recruitment highly variable. Thus, to be applied to cephalopod stocks, analytical methods require monthly estimates of biological parameters and of catch structure. Cohort analysis was successfully applied for loliginid squid stocks in the English Channel (Royer et al., 2002). In this area the cuttlefish life-cycle is different from that of common squid and a series of adjustments were necessary to adapt the cohort assessment techniques.

This study presents series of abundance indices from catch and effort data and evaluations of the stock using analytical methods, based on biological data collected since 1996. A General Linear Model is used to estimate indices of stock abundance and recruitment. A cohort analysis (Pope, 1972) is then carried out on six cuttlefish cohorts (1995–2000) on a monthly basis using catch-at-age data and the results are integrated in a Thomson and Bell model (Sparre and Venema, 1998) to simulate production and biomass of the stock and interactions between fishing fleets.

The first aim is to study abundance of the cuttlefish stock: seasonal pattern, recruitment levels and interannual variation in recruitment. Results of index estimation are compared to results of stock assessment. The second aim is to assess the impact of the fishery on the stock and to measure interactions between different groups of fishers in the exploitation of this resource. For this purpose, a diagnostic for the cuttlefish stock is provided for six cohorts that have passed completely through the fishery and a typology of interactions between métiers is proposed. Analytical models developed in this study represent useful tools to understand variability and response to fishing activity of the stock in the future.

Section snippets

Data sources

This analysis used fishery statistics extracted from French and UK national databases and biological data acquired during EU-funded research projects (“Cephvar” CT 96-1520, “Data collection for assessment of cephalopod stocks” CFP99-063 and “Cephstock” O5CA-2002-00962).

National databases were interrogated to obtain monthly records of total cuttlefish landings (kg) and of “catch and effort” data for selected fishing gears. English and Welsh data (Centre of Environment, Fisheries and Aquatic

Catch structure

Estimated monthly catches, by number, of cuttlefish (Fig. 3) varied from 0.1 to 6 million individuals over the studied period (October 1995 and June 2002) with a peak in October 2000. When catches were high in autumn and spring, 2-year-classes were exploited simultaneously, corresponding to three microcohorts in autumn and four microcohorts in spring. In summer (July and August), the stock was composed of only one cohort (two microcohorts). This structure illustrates the species’ life-cycle,

Quality of stock assessment and limits

Quality of stock assessment can be influenced by several factors, especially uncertainties in input parameters of models, such as natural mortality, catch at age, or terminal fishing mortality. However, these potential errors are generally not amplified by the model (Pope, 1972, Pelletier, 1990). Sensitivity analysis carried out on these assessments showed that errors in input terminal fishing mortality had little influence on recruitment estimates. This is just another example of the

Acknowledgements

Support for this research was provided by EU funded research programmes (AIR CT 92-0573, FAIR CT 96-1520 and Q5CA-2002-00962) and also by the “Conseil Régional de Basse Normandie”.

References (50)

  • I.A.G. Young et al.

    Application of depletion methods to estimate stock size in the squid Loligo forbesi in Scottish waters (UK)

    Fish. Res.

    (2004)
  • Anon., 2003. Working group on cephalopod fisheries and life history. Living Resources Committee, ICES CM 2003/G:02, p....
  • V. Bettencourt et al.

    Age studies based on daily growth increments in statoliths and growth lamellae in cuttlebone of cultured Sepia officinalis

    Mar. Biol.

    (2001)
  • S.V. Boletzky

    Sepia officinalis

  • E. Boucaud-Camou et al.

    The migration of the cuttlefish (Sepia officinalis) in the English Channel

  • N. Boucaud-Camou et al.

    The sexual cycle of Sepia officinalis L. from the Bay of Seine

  • P.R. Boyle
    (1983)
  • P.R. Boyle et al.

    Cephalopod population: definition and dynamics

    Phil. Trans. Roy. Soc. Lond. Ser. B.

    (1996)
  • J.K.T. Brodziak et al.

    A method to assess squid fisheries in the north-west Atlantic

    ICES J. Mar. Sci.

    (1993)
  • Caddy, J.F., 1983. The cephalopods: factors relevant to their population dynamics and to the assessment and management...
  • J.F. Caddy

    Modelling natural mortality with age in short-lived invertebrate populations: definition of a strategy of gnomonic time division

    Aquat. Living Resour.

    (1996)
  • M. Chadwick et al.

    L’analyse de capture et d’effort

  • Denis, V., 2000. Variations spatio-temporelles d’abondance des céphalopodes exploités depuis les côtes atlantiques...
  • V. Denis et al.

    Spatio-temporal analysis of commercial trawler data using General Additive models: patterns of Loliginid squid abundance in the north-east Atlantic

    ICES J. Mar. Sci.

    (2002)
  • C. Dintheer et al.

    Les grands métiers de la Manche. Reflexions et propositions pour la conservation de la ressource et la gestion des pêcheries

    Pêche Maritime

    (1995)
  • Cited by (33)

    • A Bayesian two-stage biomass model for stock assessment of data-limited species: An application to cuttlefish (Sepia officinalis) in the English Channel

      2017, Fisheries Research
      Citation Excerpt :

      Calculation of growth of 0+ group might be biased because of micro-cohort issues. For example, Royer et al. (2006) indicates the presence of two micro-cohorts of cuttlefish in the English Channel, with a first recruitment around October, and a second around April. As the CGFS takes place in October, the mean growth calculated for group 0 animals might be biased for years when there were two micro-cohorts: only the first micro-cohort would be represented in the data of age class 0 in year t, whereas both micro-cohorts would be represented for age class 1 in year t + 1.

    • Environmental effects on cephalopod population dynamics: Implications for management of fisheries

      2014, Advances in Marine Biology
      Citation Excerpt :

      Further analysis suggested the introduction of interindividual growth variability into the cohort analysis (Challier et al., 2006a). VPA was also applied as a trial to the English Channel cuttlefish stock (Royer et al., 2006) to estimate fishing impact on the studied cohorts and interactions between fishing fleets involved in the cuttlefish exploitation in the English Channel (mainly trawlers and trappers). As in the loliginid squid case, estimating age classes using statolith rings is feasible until the age of 240 days (Bettencourt and Guerra, 2001) but time-consuming and was mainly used to explore the recruitment period (Challier et al., 2002, 2005a).

    View all citing articles on Scopus
    View full text