ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on July 3, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(6):1124-1135; doi:10.1093/icesjms/fsm089
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Fluctuations and forecasts in the fishery for queen scallops (Aequipecten opercularis) around the Isle of Man
1 Port Erin Marine Laboratory, University of Liverpool, Port Erin, Isle of Man IM9 6JA, British Isles
2 Marine Conservation Society, Unit 3, Wolf Business Park, Alton Road, Ross-on-Wye HR9 5NB, UK
Correspondence to B. D. Beukers-Stewart: tel: +44 1989 561584; fax: +44 1989 566815; e-mail: bryce{at}mcsuk.org
Vause, B. J., Beukers-Stewart, B. D., and Brand, A. R. 2007. Fluctuations and forecasts in the fishery for queen scallops (Aequipecten opercularis) around the Isle of Man. – ICES Journal of Marine Science, 64: 1124–1135.The annual success of the queen scallop fishery around the Isle of Man in the northern Irish Sea is dependent on the strength of recruitment. We examined data from surveys and commercial logbooks on the annual density of spat, juvenile, and adult queen scallops in the fishery between 1982 and 2002. These were used to examine past population and fishery trends and the potential for formulating a predictive model for the fishery. The results were highly variable on both temporal and spatial scales, but there were some general trends. Density appeared to have been relatively stable during the 1980s, declined sharply from the early to mid-1990s, then recovered to produce relatively good catch rates thereafter. There was no relationship between spat settlement and the subsequent density of juveniles or adults in stock surveys or with commercial catch rates. However, within the stock surveys, there were three different significant relationships between cohort densities over time. Additionally, there was a significant relationship between the density of 1-year-olds caught on the surveys and commercial catch rates the following year. Monitoring juvenile queen scallop density would therefore allow prediction of recruitment and fisheries variations at least 1 year in advance, allowing perhaps for more effective management, including reducing the fluctuations in the fishery and helping to ensure long-term sustainability.
Keywords: Aequipecten opercularis, catch predictions, fisheries management, recruitment variation, stock assessment
Received 25 July 2005; accepted 6 May 2007; advance access publication 3 July 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Fluctuations in the abundance of fish and shellfish populations are often attributed to variations in recruitment (Sissenwine, 1984). This is generally caused by several interacting factors (Rothschild, 2000), most commonly the size of the spawning stock (Ricker, 1954; Beverton and Holt, 1957; Shepherd, 1982) and environmental conditions (Caputi, 1993; Neill et al., 1994; Hofmann and Powell, 1998; Le Pennec et al., 2003). It is important that such variation in recruitment be accounted for in managing exploited stocks, even when the cause of the variation cannot be identified. An ability to predict the level of recruitment, defined here as the abundance of individuals entering the fishery, contributes to more effective fisheries management (Smith, 1993). Such prediction would be particularly valuable for fisheries that are heavily dependent on the strength of the recruiting year class, such as many shellfish stocks.
The queen scallop (Aequipecten opercularis) is a commonly exploited species in the Northeast Atlantic (Brand, 1991), and is common around the British Isles (Mason, 1983; Ansell et al., 1991). It is rare to find scallops older than 6 years (Brand et al., 1991a), possibly because of the combined effect of cumulative fishing mortality and the onset of senescence (Allison and Brand, 1995), but in this time, they can grow to 90 mm shell height (Brand et al., 1991a). The most valuable fishery for the species is in the Irish Sea; 9286 t (45% of the total landings in the Northeast Atlantic) were taken from the region in 2001 (FAO, 2003). A significant part of the fishery operates out of the Isle of Man, with an average of >2300 t of queen scallops landed there each year between 1982 and 2002 (Isle of Man Department of Agriculture, Fisheries and Forestry statistics). The Manx queen scallop fishery operates mainly during summer (June–October inclusive), and is prosecuted with two different types of fishing gear, dredges or trawls. Both are size-selective to target animals >55 mm shell height, because of the economics of processing (Brand et al., 1991a). During the fishing season, there is a change in the size structure of the population, caused by rapid growth of scallops in summer. This can produce a change in the age structure of the catch as the recruiting cohort, generally 2-year-olds, often attain 55 mm in the later part of the season and so become vulnerable to fishing (Allison, 1993). The fishery is dominated by scallops 2–4 years old (Brand et al., 1991a; Allison and Brand, 1995) and the few age classes in the exploited population dictates that the success of the fishery each year is very dependent on the strength of the incoming year class. For fisheries such as this, recruitment variation is a major contributor to temporal variations in the commercial catch.
Scallop stocks have a well established reputation for being temporally and spatially variable, and the main causes have been summarized into three groups; recruitment variability, catastrophic mortality, and the longevity of the species (short-lived species have no buffer zone if there is a period of poor recruitment, making them more vulnerable to recruitment failure) (Orensanz et al., 1991). The rapid early growth, short lifespan and high motility of A. opercularis led Mason (1983) to state that queen scallop stocks needed no protection and that the best fishing strategy was to "fish them hard before some other predator gets them". Similarly, Hancock (1979) considered recruitment for some scallop populations to be so unreliable that for management purposes, they should be regarded as a non-renewable resource.
Many previous studies of fish (Helle et al., 2000; Nash and Geffen, 2000; Arnott and Ruxton, 2002) and shellfish (Chícharo and Chícharo, 2001; Beukers-Stewart et al., 2003; Caputi et al., 2003) have documented good relationships between the abundance of cohorts at different life history stages. In the northern Irish Sea, Beukers-Stewart et al. (2003) recorded a strong link between settlement of great scallop (Pecten maximus) spat and commercial catch per unit effort (cpue) 4–5 years later, and a previous study on queen scallops in Scottish waters suggested a positive relationship between spat settlement and the subsequent abundance of juveniles and adults (Fraser, 1991).
Our study brings together 20 years of data from three monitoring programmes that were designed to assess the queen scallop populations and their fishery around the Isle of Man. These programmes measured the density of queen scallops at different life history stages. Year classes were sampled for the first time as spat settled out of the plankton onto spat collectors, and stock surveys caught juveniles and adults which supported length frequency analysis as a basis for determining relative cohort densities. Additionally, commercial fishers contributed catch and effort data in the form of logbooks. All these data were combined to investigate the potential for developing a predictive model for the fishery.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Data on spat settlement, on stock structure from the surveys, and on commercial catch and effort from logbooks were analysed for four fishing grounds around the south and east of the island (Figure 1), for various periods between 1982 and 2002.
|
Spat settlement
Data from a long-term (1975–2001) monitoring programme of pectinid spat settlement around the Isle of Man (Brand et al. 1991b; Beukers-Stewart et al., 2003) were examined, and information on the annual spat settlement of A. opercularis at four sites between 1991 and 2001 was selected for analysis: Bradda, Bay Fine, Bay Stacka, and Niarbyl (Figure 2). Three lines of spat collectors were deployed at each site at the beginning of each settlement season (normally June), with a spat bag placed at four different depths above the seabed along each line. The number of spat settled in each bag at the end of the season (normally September) was then counted to calculate the mean number of spat per bag at each site (see Beukers-Stewart et al., 2003, for a more detailed description of the method). Annual variation in spat settlement was investigated statistically by a one-way analysis of variance (ANOVA), using just Bradda and Bay Fine as replicates because those sites had the most complete data series. There were no data available on replication within these sites. As the design was balanced, the ANOVA was robust to any departure from homogeneity of variance (Underwood, 1997). A Student–Newman–Keuls (SNK) test was conducted to identify significant differences between years.
|
Stock surveys
Fisheries-independent surveys were conducted between 1992 and 2002 on an average of five of the eight queen scallop fishing grounds around the Isle of Man. In this study, the fishing grounds South Port St Mary (South PSM), East Douglas, Southeast (SE) Douglas, and Laxey (Figure 1) were analysed because they had the most complete datasets. The surveys were conducted in June and October (the start and the end of the peak queen scallop fishery). The RV "Roagan", a 24 m converted beam trawler, towed a gang of four spring-loaded Newhaven queen scallop dredges (Chapman et al., 1977; Mason, 1983). Each dredge was 0.76 m wide, had ten teeth 60 mm long, and belly rings 55 mm internal diameter. Each tow was for
2 nautical miles (3.70 km) at an average speed of 2.5 knots (4.63 km h–1), thus lasting 45–50 min. Each ground was surveyed by conducting three or four replicate tows. A differential global positioning system (dGPS) and Microplot software (Sea Information Systems, Aberdeen) were used to ensure that the same area of seabed was targeted during each survey. All A. opercularis caught were counted, and the shell height of a subsample (generally one-third of the catch) was measured to the nearest millimetre. Shell height was defined as the maximum distance between the dorsal hinge and the ventral margin (Seed, 1980). No attempt was made to age A. opercularis at this point because interpretation of the annual rings can be problematic (Kaiser et al., 2000). Instead, year classes were identified using Bhattacharya's method of length frequency analysis (Valencia, 1988; Sparre et al., 1989). The first three year classes tend to form clear, normally distributed modes; thereafter, growth slowed considerably and it was difficult to differentiate between year classes after 3 years of age. The relative densities of year classes 1, 2, 3 and 4+ were calculated and expressed as the mean number of A. opercularis per 100 m2 of swept seabed, using each tow as a replicate. The efficiency of the dredges for catching queen scallops was not known, so it was not possible to estimate actual density.
Temporal and spatial variation in the relative density of A. opercularis was investigated. The data were tested for homogeneity of variances (Cochran's C-test) and transformed where necessary. Two-way ANOVAs (followed by SNK tests when results were significant) were conducted to examine the effect of year and month (June or October) on density estimates at each ground.
Commercial logbooks
A voluntary logbook scheme for northern Irish Sea scallop fishers was started in 1982 (Brand and Allison, 1994; Beukers-Stewart et al., 2003), and an average of 30% of the fleet participated in the scheme. Fishers recorded the type and size of the gear used, the landed catch (number of standard size bags of queen scallops), the time spent fishing, and the location of fishing according to a grid map of 5 x 5 nautical mile squares (Figure 1). This small spatial scale is particularly suitable for semi-sedentary, patchily distributed species such as queen scallops. In some cases, the larger fishing grounds consisted of several grid squares. Using these data, commercial cpue (number of bags of queen scallops caught per metre of gear width per hour) was calculated for each ground per year. Aequipecten opercularis are fished commercially with both dredges and trawls. The dredges are the same design as those used in the stock surveys and the fishers recorded the width (of the tooth bar) and the number of dredges used. Trawls are towed singly and were measured across the header rope; however, the actual area of seabed covered was less than that owing to the flexibility of the system. For this reason, absolute values of cpue for dredges and trawls were not directly comparable, although it was valid to compare trends in the two fisheries. Dredges and trawls are towed at a consistent speed (2.5–3 knots) throughout the fleet, so commercial boats are comparable with each other, and with our surveys. There has been little or no change in fishing gear design, average boat size, or engines over the course of this study (unpublished data), so catch rates were also considered to be comparable over time.
Catch prediction
The relationships between the abundance of A. opercularis at different life history stages were explored through regression analysis, using combinations of year and ground as replicates. The relationships between mean annual spat settlement (at the Bradda and Bay Fine sites) and commercial cpue 2 and 3 years later were tested for dredges and trawls separately. Data from all available sites and years were used as replicates in this analysis. Changes in the densities of year classes over time were calculated from the June survey data. To ascertain whether the stock survey data were representative of commercial cpue, the density of 2-year-olds and older was compared with commercial cpue for the same year, for both dredges and trawls. Further, the relationship between the density of 1-year-olds from the stock surveys and commercial cpue 1 year later was examined for both gear types. Using these relationships, cpue was predicted from the density of 1-year-olds for the period studied and compared with the actual cpue.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Spat settlement
Between 1991 and 2001, mean spat settlement of A. opercularis around the Isle of Man was 228.7 (s.e. 65) spat per collector. There was significant temporal variation in spat settlement between years (one-way ANOVA; F1,7 = 202.45, p < 0.001) caused mainly by the significantly better settlement in 1993 and 1994 (significant SNK test, Figure 3). Maximum settlement over the 11-year period of study (mean number of spat per bag = 1432.73) was at Bay Fine in 2001, but the collectors were lost from Bradda that year, so the data for that year could not be included in the analysis. There was considerable spatial variation, e.g. in 1998, when there was a 35-fold difference between the mean number of spat per collector at Bay Fine and Bay Stacka. Spat settlement at Bradda and Bay Fine was usually fairly similar and consistently greater than settlement at Bay Stacka and Niarbyl.
|
Stock surveys
There was great temporal and spatial variation in the relative density of A. opercularis during the stock surveys (Figure 4). For the period 1992–1996, the density of queen scallops on all grounds varied between 2 and 5 per 100 m2. In 1997 and 1999, density was much higher, except at SE Douglas, which was poorly sampled (note that there was no sampling at any site during 1998). Queen scallop density first increased at East Douglas in 1997, then remained relatively high until the end of the study in 2002. At Laxey, density increased in 1999 and remained relatively high until 2002, when it fell back to the previous levels of <5 per 100 m2. In June 2000, South PSM had the greatest density of queen scallops for the study period, 20.5 per 100 m2.
|
Results of a two-way ANOVA revealed significant interaction between year and month (June or October) for South PSM, East Douglas, and Laxey (Table 1). An SNK test found a small (1.8–3.4 queen scallops per 100 m2) but significant difference between June and October results for 1992, 1994, and 1995 (Table 2). However, the significant differences found in 1997, 1999, and 2000 were much greater, relating to a mean difference of 7.6 queen scallops per 100 m2. The trends in the later part of this study indicate that more significant differences may have been found if a more complete data series had been available. Always, density in October was significantly greater than in June.
|
|
Commercial logbook analysis
On the four commercial grounds researched in this study, dredges are used at South PSM, East Douglas, and SE Douglas, and trawls at East Douglas, SE Douglas, and Laxey (Figure 1). Substratum, time of year, and personal preference determine the type of gear fishers use (Brand and Allison, 1987). For the 20-year monitoring period, commercial cpue fluctuated widely for both dredge and trawl fisheries (Figure 5). For the first 10 years, dredge cpue fluctuated around the overall mean (0.5 bags m–1 h–1), then began to drop in the early 1990s and bottomed out at 0.17 bags m–1 h–1 in 1995. The fishery recovered well from that year, and in 2000, reached a high of 0.85 bags m–1 h–1. The trawl fishery was relatively productive in the first 10 years, rarely dropping below the mean of 0.12 bags m–1 h–1. Like the dredge fishery, trawl cpue fell in the early 1990s but took longer to recover, not moving above the long-term mean cpue until 2001.
|
Spat settlement predictions of stock survey density and commercial cpue
Relationships between spat settlement and the density of 1- and 2-year-old queen scallops in the June and October stock surveys 1 and 2 years later were investigated using regression analysis. No significant relationships were found with individual fishing grounds or with mean density for all grounds (Table 3). The increase in queen scallop density at East Douglas in 1997 may have been a result of the excellent spat settlement in 1994, but the dramatic increase in scallop density at South PSM and Laxey in October 1999 was not seen in the spat settlement of 1997, as would have been expected (Figures 3 and 4). Similarly, there were no relationships between spat settlement and dredge or trawl cpue with time-lags of 2 and 3 years for each ground individually or overall (Table 3). Commercial cpue was at its lowest in 1994 and 1995 and the lowest mean spat settlement of 38 spat per bag was recorded 3 years earlier, in 1992. However, the high cpue values in 2000 and 2001 were not detected in the spat settlement in 1998, which was below average (Figures 3 and 5).
|
Stock density predictions within surveys
Regression analysis showed strongly significant linear relationships between the densities of cohorts over time (Figure 6) using data from all grounds and years. A 1-year time-lag was first established for 1-year-olds against 2-year-olds (r2 = 0.81) and 2-year-olds against 3-year-olds (r2 = 0.75). Results were similar for a 2-year time-lag between 1- and 3-year-olds (r2 = 0.87). In all these results, p < 0.001. Perusal of Figure 6 shows that the significant regressions may have been unduly influenced by a single point representing the high density of a single cohort (1-year-olds in 1999, 2-year-olds in 2000, and 3-year-olds in 2001). However, even without this point, all relationships were significant (without the outlier the relationships between relative cohort density over time were: 1-year-olds against 2-year-olds y = 8.49x + 0.41, r2 = 0.39, p < 0.001; 2-year-olds against 3-year-olds y = 0.42x + 0.98, r2 = 0.22, p = 0.016; 1-year-olds against 3-year-olds y = 8.07x + 0.68, r2 = 0.39, p = 0.003). Therefore, within stock surveys, the densities of 1- or 2-year-olds could be used to predict the density of 2- or 3-year-olds up to 2 years in advance.
|
Stock survey density predictions of commercial cpue
Aequipecten opercularis first enter the fishery during the fishing season when they are 2 years of age. The density of A. opercularis
2-year-olds in the June stock surveys was tested against commercial cpue the same year and strongly significant linear relationships were found (dredge r2 = 0.81, p < 0.001; trawl r2 = 0.36, p = 0.014), showing that the stock surveys were representative of commercial catches (Figure 7). Following this, the density of 1-year-olds from the surveys was tested against commercial cpue 1 year later. Again, strong linear relationships were found (dredge r2 = 0.71, p < 0.001; trawl r2 = 0.42, p = 0.016), allowing commercial cpue to be predicted 1 year in advance. Using these predictive relationships, commercial cpue values for 10 years of the study period were calculated and compared with the actual cpue recorded in the logbooks (Figure 8). The predicted density of 2-year-olds from the stock surveys paralleled actual cpue reasonably closely for both fisheries, indicating that a large proportion of the catch is 2-year-olds, the recruiting year class. The lack of survey data for June 1998 prevented predictions being made for 1999. In 2000, there was exceptionally large recruitment which resulted in very high catch rates; this was overestimated for the dredge fishery but well predicted for the trawl fishery. Catch rates were underestimated for both fisheries in 2001 (see below for a possible explanation of this finding).
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Studies of invertebrate recruitment suggest that variation is often independent of the abundance of spawners (particularly those with high fecundity) and mainly influenced by environmental variability (Hancock, 1973; Drinkwater and Myers, 1987; Caputi, 1993). There have been many previous studies on recruitment and the affect of abiotic factors, e.g. temperature (Dickie, 1955; Fogarty, 1988; Mackenzie and Köster, 2004), salinity (Nell and Holliday, 1988; Laing, 2002), suitability of habitat (Stokesbury and Himmelman, 1995), and biotic factors, e.g. food availability (Jackson et al., 1995), indirect fishing mortality (Shepard and Auster, 1991), and predator abundance and competition (Thouzeau, 1991). All these factors vary spatially and temporally, so recruitment is often inconsistent.
Variations in recruitment are often a major contributor to fluctuations in pectinid fisheries (Orensanz et al., 1991) and the absence of stock–recruitment relationships in scallop populations has been well documented (Caddy, 1979; Mason, 1983; Sinclair et al., 1985; Orensanz et al., 1991). For these reasons, we decided to monitor and analyse the Isle of Man A. opercularis populations using relationships between early and later life history stages. However, when managing fisheries, it is dangerous to assume that reducing the number of spawners will not affect recruitment, and there are several documented incidences of recruitment-overfishing of bivalve populations (Dredge, 1988; Young and Martin, 1989; Peterson, 2002). Most likely, because environmental factors play such an important role in recruitment, dependence on the size of the spawning stock is disguised.
The Irish Sea (ICES fishing Area VIIa) has consistently yielded the largest annual landings of A. opercularis in Europe, since the fishery began in 1969 (Ansell et al., 1991; FAO, 2003). However, the productivity of the fishery has always been very variable. Large peaks and troughs in total annual landings of queen scallops were recorded for the first 20 years of the fishery (Mason, 1983; Ansell et al., 1991), but it is not known how changes in fishing effort may have influenced the patterns. Following from this, our more recent 20 years of logbook data plus 10 years of fisheries-independent surveys show that there have been great fluctuations in commercial cpue and relative population density around the Isle of Man.
To interpret and utilize the results of stock assessments, it is important to consider the efficiency and selectivity of the fishing gear being used, as well as any variation (Hilborn and Walters, 1992; Fifas and Berthou, 1999). An investigation using spring-toothed scallop dredges for P. maximus in the northern Irish Sea estimated 30–40% efficiency for catching scallops of legal size (Beukers-Stewart et al., 2001), but the efficiency of the dredges used for catching for A. opercularis is unknown. The fishing gear is designed to target scallops >55 mm, and it is assumed that catch efficiency increases with size and therefore with age (Fifas and Berthou, 1999; Beukers-Stewart et al., 2001). Consequently, catch efficiency should be constantly increasing during the course of a fishing season commensurate with growth. Additionally, at the start of the season in June, the fishery consists mainly of 3-year-olds and older. During the fishing season, the recruiting 2-year-old year class grows to target size, becoming increasingly vulnerable to the fishing gear and therefore available to the fishery. However, all this is complicated by the seasonal swimming behaviour of queen scallops, which also affects catchability (Yonge, 1936; Moore and Trueman, 1971; Chapman et al., 1979). Jenkins et al. (2003) found a strong correlation between seawater temperature and A. opercularis swimming activity. They calculated that up to 42% of queen scallops (>55 mm) in the dredge path evaded capture by swimming over the dredge mouth, and showed too that larger scallops have a lesser propensity to swim than smaller ones.
Our stock surveys were conducted at the start and the end of the peak queen scallop fishing season (June and October, respectively). At those times, the swimming behaviour of queen scallops is similar (Jenkins et al., 2003). Stock density is likely to be consistently underestimated from the June surveys because of lesser catching for the recruiting year class, which ultimately constitutes the main fished stock. During the course of each fishing season, the stock undergoes many changes as it is subject to a complex combination of effects: growth, recruitment, changes in capture efficiency, and fishing mortality (both direct and indirect) (Allison, 1993). By the second stock survey in October, most of the recruiting year class is larger than the minimum size targeted, so will be better represented in the catch. The October surveys therefore provide a better estimate of total stock density, but the recruiting year class will have been subject to direct and indirect fishing mortality in the fishing season leading up to that month. These interrelated factors make it difficult to estimate the actual size of the recruiting year class. Therefore, although the June surveys operate at a lesser efficiency for the recruiting year class, they are more useful for examining variations in the relative strength of recruitment.
Unlike some other scallop fisheries (Beukers-Stewart et al., 2003), it seems that the northern Irish Sea queen scallop fleet may not always fish down a ground until it becomes subeconomic. Evidence for this comes from the consistently greater densities in October (the end of the season) than in June, suggesting that recruitment to the fishable stock is generally greater than fishing mortality. The changes occurring during the fishing season make it very difficult to quantify the level of exploitation except perhaps by tagging (Walters and Martell, 2004). Allison and Brand (1995) conducted a mark-recapture experiment on A. opercularis for the 1989 fishing season on the East Douglas fishing ground. Their results yielded mean densities of commercial sizes of queen scallops for the ground of 0.47 m2 (although they noted that the entire area would not support commercially exploitable populations, so actual densities within the fished areas were likely to be higher) and overall exploitation rate to be 25.4% for the fishing season. Similar to this study, they found that 2- and 3-year-olds dominated the fishery, in their case 2-year-olds accounting for 42%, and 2- and 3-year-olds combined accounting for 78% of the total catch.
The natural settlement of A. opercularis on spat collectors was spatially, but more so temporally, highly variable. In some instances, it provided an indication of a strong or weak recruitment to the fishery (1994 and 1992, respectively), but in others it did not (1997 and 1998). The inconsistency in this relationship is likely to have been attributable to a combination of factors including: temporal and spatial variability in local hydrographic conditions that determine the supply and retention of larvae (Zhang, 1996); temporal and spatial variability in the survival of spat (both in the collectors and on the seabed); and practical limitations on where the collectors could be located (Beukers-Stewart et al., 2003). A better understanding of the factors influencing settlement and early survival of queen scallops could allow a more conversant spat collection programme to be designed. This may allow predictions for the queen scallop fishery to be made 2 years (or more) in advance and would be a valuable tool for managing the fishery.
The fisheries-independent stock surveys showed more promising results. The density of 1-year-olds could predict the density of 2-year-olds the following year, and the density of 3-year-olds 2 years later, and the density of 2-year-olds could also be used to predict the density of 3-year-olds the following year. The exceptional recruitment in 1999/2000 at the South PSM fishing ground had a strong influence on these regressions. This is very significant because it indicates that a good recruitment of juveniles (1-year-olds) does persist/survive to become adults that enter and ultimately support the fishery. Moreover, without this outlying point, the relationships were still significant (although not as strong). In all forecast calculations, June (rather than October) stock survey data were used because they were better for examining recruitment (as discussed earlier), and provided earlier predictions, which is obviously more desirable.
Fisheries-independent stock surveys usually include random sampling to estimate population density (Hilborn and Walters, 1992). The results often differ from commercial cpue estimates because fishers target high-density patches of scallops of marketable size. In this study, the stock surveys were designed to target productive areas, to be representative of the commercial catch. The strong relationships we observed between the estimated density of 2-year-olds and commercial cpue indicate that stock surveys were indeed appropriate for both dredge and trawl fisheries. In addition, the relationship between density of 1-year-olds and cpue the following year was good, allowing catch predictions to be made and further supporting the notion that 2-year-olds generally dominate the fishery. The cluster of points around zero was because of the generally low capture efficiency for 1-year-olds in June. However, when a cohort of 1-year-olds was both abundant and of larger than average size at age, catch rates increased. These cohorts were also likely to make bigger contributions to the following year's fishery, so the density of 1-year-olds is assumed to be a good indicator of future catch rates. With good evidence that stock surveys are representative of commercial catch rates and that the fishery mainly consists of 2-year-olds, the relationship between the density of 1-year-olds and the density of 2-year-olds in the next year's survey was used to predict commercial cpue for the study period. There was very little difference between actual and predicted cpue for both fisheries between 1993 and 1998. In 2000, cpue was overestimated for the dredge fishery but well estimated for the trawl fishery, but in 2001, cpue was underestimated for both fisheries. In 2000, there was exceptionally good recruitment on the South PSM fishing ground, and the densities at East Douglas and Laxey were also very high (SE Douglas was not sampled). This boom flooded the market, and the processors set landing restrictions (unpublished data; see also Brand et al., 1991a). As a consequence, high densities of queen scallops remained at the end of the season, especially at South PSM. Evidence for this is also provided in the stock surveys of October 2000 and June 2001, when the cohort was 3 years old. This simple model is only predicting the density of 2-year-olds (normally the dominant year class in the catch), but in 2001, the exceptionally large 3-year-old cohort was likely to have made a massive contribution to the fishery. Therefore, it seems that, when there is an exceptionally large recruitment, one year class may support and dominate two fishing seasons. This is a likely explanation for the underestimation of commercial catches in 2001, but it must be stressed such strong recruitments are rare. The predictive model developed in this study assumes that the fishery is entirely reliant on recruiting 2-year-olds, but 3-year-olds and above are an important subsidiary component of the commercial catch (especially at the start of the season). By not incorporating older year classes into the model, the predictions take a more conservative approach, but as more information becomes available, the model could be refined to improve the predictions.
In any fishery, it is of obvious benefit to predict recruitment as early as possible. To do this in the Manx fishery for queen scallops, increased understanding of the population dynamics of A. opercularis is required. A problem in many fisheries, including that for the northern Irish Sea queen scallop, is that for all grounds the location of the parent stock is uncertain. This makes it particularly difficult to define the stock–recruitment relationship. There is conflicting genetic evidence about A. opercularis populations being discrete and self-sustaining (Macleod et al., 1985; Sinclair et al., 1985; Lewis and Thorpe, 1994). It is widely believed that oceanographic features play a major role in the presence or absence of scallop populations and recruitment to those populations (Orensanz et al., 1991; Young et al., 1992). To predict the dispersal of pelagic larvae, models of water movement and larval behaviour need to be applied to the stocks (McShane et al., 1988; Bartsch et al., 1989; Hill, 1990; Tremblay et al., 1994; Bradbury and Snelgrove, 2001). Other areas recommended for further study include the influence of abiotic and biotic factors on growth and survival, spatial and temporal investigations into the age structure and growth rates of the populations, and the development of a model to estimate catch efficiency incorporating all the confounding factors mentioned earlier.
This study suggests that the northern Irish Sea queen scallop fishery is often heavily reliant on recruiting 2-year-olds and less so on 3-year-olds, so making the fishery potentially vulnerable to recruitment-overfishing. As queen scallops are broadcast spawners, a decrease in density is likely to rapidly reduce fertilization efficiency (Stoner and Ray-Culp, 2000). The population size below which the stock cannot recover is unknown, and because A. opercularis are aggregative (Brand, 1991), determining this threshold will be very complicated (Lundquist and Botsford, 2004). Interestingly, a spat collection study of Argopecten irradians found that settlement (and subsequent survival) was sometimes high at sites of low adult spawner abundance, whereas collectors at sites with high adult abundance always received good settlement (Ambrose et al., 1992). One potential management strategy for A. opercularis (if sources of larval supply can be identified) is the use of closed areas. In areas of no fishing disturbance, it is likely that there will be an increase in population density (Halpern, 2003) and habitat complexity (Kaiser et al., 2000; Bradshaw et al., 2001, 2003). These combined effects are likely to increase reproductive output and potentially recruitment to surrounding areas open to fishing (Beukers-Stewart et al., 2005).
The persistence of the Isle of Man fishery for queen scallops since 1969 and the recent upward trend in indicators of stock size and commercial catch rate suggests that current effort levels in the fishery are sustainable. However, this study joins a growing number, particularly among invertebrate fisheries, that demonstrate the potential for using predictive models to aid in fisheries management. The results from robust predictive models could be built into decision-rule frameworks for managing these fisheries, allowing for catch and effort levels, and season start times and durations, to be varied according to the strength of incoming year classes. These measures could not only assist in ensuring the long-term viability of the fisheries, but also increase their productivity and mitigate their effects on the wider ecosystem.
|
Acknowledgements |
|---|
Funding for this work was provided by the Department of Agriculture, Fisheries and Forestry, Isle of Man Government. We are especially grateful to Richard Nash, Stuart Jenkins, and Antonio Valencia, who provided invaluable statistical advice and comments on early versions of the manuscript. Comments and advice from Audrey Geffen and two anonymous referees also improved the manuscript considerably during the review process. We are indebted to the boat crews of the RV "Roagan" and RV "Sula" and all those who helped collect the data over the years, especially Matthew Mosley, Graham Hughes, Kate Prudden, Dave Pennington, and Ulli Wilson. Ross Huggett, John Stead, and Clare Goodwin processed many of the spat collectors.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Allison E. H. The dynamics of exploited populations of scallops (Pecten maximus L.) and queens (Chlamys opercularis L.) in the North Irish Sea. (1993) UK: University of Liverpool. 254. PhD thesis.
Allison E. H., Brand A. R. A mark-recapture experiment on queen scallops, Aequipecten opercularis, on a north Irish Sea fishing ground. Journal of the Marine Biological Association of the UK (1995) 75:323–335.
Ambrose W. G., Peterson C. H., Summerson H. C., Lin J. Experimental tests of factors affecting recruitment of bay scallops (Argopecten irradians) to spat collectors. Aquaculture (1992) 108:67–86.[CrossRef][Web of Science]
Ansell A. D., Dao J-C., Mason J. Three European scallops: Pecten maximus, Chlamys (Aequipecten) opercularis and C. (Chlamys) varia. In: Scallops: Biology, Ecology and Aquaculture—Shumway S. E., ed. (1991) Amsterdam: Elsevier. 715–751.
Arnott S. A., Ruxton G. D. Sandeel recruitment in the North Sea: demographic, climatic and trophic effects. Marine Ecology Progress Series (2002) 238:199–210.[Web of Science]
Bartsch J., Brander K., Heath M., Munk P., Richardson K., Svendsen E. Modelling the advection of herring larvae in the North Sea. Nature (1989) 340:632–636.[CrossRef]
Beukers-Stewart B. D., Jenkins S. R., Brand A. R. The efficiency and selectivity of spring-toothed scallop dredges: a comparison of direct and indirect methods of assessment. Journal of Shellfish Research (2001) 20:121–126.[Web of Science]
Beukers-Stewart B. D., Mosley M. W. J., Brand A. R. Population dynamics and predictions in the Isle of Man fishery for the great scallop, (Pecten maximus, L.). ICES Journal of Marine Science (2003) 60:224–242.
Beukers-Stewart B. D., Vause B. J., Mosley M. W. J., Rossetti H. L., Brand A. R. Benefits of closed area protection for a population of scallops. Marine Ecology Progress Series (2005) 298:189–204.[CrossRef][Web of Science]
Beverton R. J. H., Holt S. J. On the dynamics of exploited fish populations. (1957) London, UK: Ministry of Agriculture Fisheries and Food. 533. Fisheries Investigations Series 2, 19.
Bradbury I. R., Snelgrove P. V. R. Contrasting larval transport in demersal fish and benthic invertebrates: the roles of behaviour and advective processes in determining spatial pattern. Canadian Journal of Fisheries and Aquatic Sciences (2001) 58:811–823.
Bradshaw C., Collins P., Brand A. R. To what extent does upright sessile epifauna affect benthic biodiversity and community composition? Marine Biology (2003) 143:783–791.[CrossRef]
Bradshaw C., Veale L. O., Hill A. S., Brand A. R. The effect of scallop dredging on Irish Sea benthos: experiments using a closed area. Hydrobiologia (2001) 465:129–138.[CrossRef][Web of Science]
Brand A. R. Scallop ecology: distributions and behaviour. In: Scallops: Biology, Ecology and Aquaculture—Shumway S. E., ed. (1991) Amsterdam: Elsevier. 517–584.
Brand A. R., Allison E. H. The Isle of Man fishery for the queen scallop, Chlamys opercularis (L.). (1987) 4. Sixth International Pectinid Workshop, Menai Bridge, 9–14: Wales, UK. 11. 1987.
Brand A. R., Allison E. H. Estimating abundance on north Irish Sea scallop, Pecten maximus (L.), fishing grounds. Bourne N. F., Bunting B. L., Townsend L. D., eds. (1994) Proceedings of the Ninth International Pectinid Workshop, 22–27 April 1993. 121–129. Nanaimo, B.C Canadian Technical Report of Fisheries and Aquatic Sciences.
Brand A. R., Allison E. H., Murphy E. J. North Irish Sea scallop fisheries: a review of the changes. In: An International Compendium of Scallop Biology and Culture—Shumway S. E., Sandifer P. A., eds. (1991a) Baton Rouge: World Aquaculture Society. 204–218.
Brand A. R., Wilson U. A. W., Hawkins S. J., Allison E. H., Duggan N. A. Pectinid fisheries, spat collection, and the potential for stock enhancement in the Isle of Man. ICES Marine Science Symposia (1991b) 192:79–86.
Caddy J. F. Long-term trends and evidence for production cycles in the Bay of Fundy scallop fishery. (1979) 175:97–108. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer.
Caputi N. Aspects of spawner recruit relationships, with particular reference to crustacean stocks—a review. Australian Journal of Marine and Freshwater Research (1993) 44:589–607.[CrossRef]
Caputi N., Chubb C., Melville-Smith R., Pearce A., Griffin D. Review of relationships between life history stages of the western rock lobster, Panulirus cygnus, in Western Australia. Fisheries Research (2003) 65:47–61.[CrossRef][Web of Science]
Chapman C. J., Main J., Howell T., Sangster G. I. The swimming speed and endurance of the queen scallop Chlamys opercularis in relation to trawling. In: Progress in Underwater Science, 4—Gamble J. C., George J. D., eds. (1979) London: Pentech Press. 57–72.
Chapman C. J., Mason J., Kinnear J. A. M. Diving observations on the efficiency of dredges used in the Scottish fishery for the scallop, Pecten maximus (L.). Scottish Fisheries Research Report (1977) 10:1–16.
Chícharo L., Chícharo M. A. A juvenile recruitment prediction model for Ruditapes decussates (L.) (Bivalvia: Mollusca). Fisheries Research (2001) 53:219–233.[CrossRef][Web of Science]
Dickie L. M. Fluctuations in abundance of the giant scallop, Placopecten magellanicus (Gmelin), in the Digby area of the Bay of Fundy. Journal of the Fisheries Research Board of Canada (1955) 12:797–857.[Web of Science]
Dredge M. C. L. Recruitment overfishing in a tropical scallop fishery? Journal of Shellfish Research (1988) 7:233–239.
Drinkwater K. F., Myers R. A. Testing predictions of marine fish and shellfish landings from environmental variables. Canadian Journal of Fisheries and Aquatic Sciences (1987) 44:1568–1573.
FAO. FISHSTAT Plus: Universal software for fishery statistical time series. (2003) version 2.3 2000.
Fifas S., Berthou P. An efficiency model of a scallop (Pecten maximus, L.) experimental dredge: sensitivity study. ICES Journal of Marine Science (1999) 56:489–499.
Fogarty M. J. Time series models of the Maine lobster fishery: the effect of temperature. Canadian Journal of Fisheries and Aquatic Sciences (1988) 45:1145–1153.
Fraser D. I. Settlement and recruitment in Pecten maximus (Linnaeus, 1758) and Chlamys (Aequipecten) opercularis (Linnaeus, 1758). In: An International Compendium of Scallop Biology and Culture—Shumway S. E., Sandifer P. A., eds. (1991) Baton Rouge: World Aquaculture Society. 28–36.
Halpern B. S. The impact of marine reserves: do reserves work and does reserve size matter? Ecological Applications (2003) 13:S117–S137.[CrossRef][Web of Science]
Hancock D. A. The relationship between stock and recruitment in exploited invertebrates. (1973) 164:113–131. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer.
Hancock D. A. Population dynamics and management of shellfish stocks. (1979) 175:8–19. Rapports et Procès-Verbaux des Réunions du Conseil International pour l'Exploration de la Mer.
Helle K., Bogstad B., Marshall C. T., Michalsen K., Ottersen G., Pennington M. An evaluation of recruitment indices for Arcto-Norwegian cod (Gadus morhua L.). Fisheries Research (2000) 48:55–67.[CrossRef][Web of Science]
Hilborn R., Walters C. J. Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty. In: Chapman and Hall (1992) New York. 570.
Hill A. E. Pelagic dispersal of Norway lobster Nephrops norvegicus larvae examined using an advection-diffusion-mortality model. Marine Ecology Progress Series (1990) 64:217–226.[Web of Science]
Hofmann E. E., Powell T. M. Environmental variability effects on marine fisheries: four case histories. Ecological Applications (1998) 8:S23–S32.[CrossRef][Web of Science]
Jackson D. L., Chisholm L. S., Merry H., O'Dor R. K. Influences of food availability, salinity, and temperature on growth response of larval sea scallops, Placopecten magellanicus. (1995) Tenth International Pectinid Workshop, Book of Abstracts, 26 April–2 May: Cork, Ireland. 99–100. 1995.
Jenkins S. R., Lart W., Vause B. J., Brand A. R. Seasonal swimming behaviour in the queen scallop (Aequipecten opercularis) and its effect on dredge fisheries. Journal of Experimental Marine Biology and Ecology (2003) 289:163–179.[CrossRef][Web of Science]
Kaiser M. J., Ramsay K., Richardson C. A., Spence F. E., Brand A. R. Chronic fishing disturbance has changed shelf sea benthic community structure. Journal of Animal Ecology (2000) 69:494–503.[CrossRef][Web of Science]
Laing I. Effect of salinity on growth and survival of king scallop spat (Pecten maximus). Aquaculture (2002) 205:171–181.[CrossRef][Web of Science]
Le Pennec M., Paugam A., Le Pennec G. The pelagic life of the pectinid Pecten maximus—a review. ICES Journal of Marine Science (2003) 60:211–223.
Lewis R. I., Thorpe J. P. Temporal stability of gene frequencies within genetically heterogeneous populations of the queen scallop, Aequipecten (Chlamys) opercularis. Marine Biology (1994) 121:117–126.[CrossRef]
Lundquist C. J., Botsford L. W. Model projections of the fishery implications of the Allee effect in broadcast spawners. Ecological Applications (2004) 14:929–941.[CrossRef][Web of Science]
Mackenzie B. R., Köster F. W. Fish production and climate: sprat in the Baltic Sea. Ecology (2004) 85:784–794.[CrossRef][Web of Science]
Macleod J. A. A., Thorpe J. P., Duggan N. A. A biochemical genetic study of population structure in queen scallop (Chlamys opercularis) stocks in the northern Irish Sea. Marine Biology (1985) 87:77–82.[CrossRef]
Mason J. Scallop and Queen Fisheries in the British Isles. In: Fishing News Books (1983) UK. 144. Farnham.
McShane P. E., Black K. P., Smith M. G. Recruitment processes in Haliotis rubra (Mollusca: Gastropoda) and regional hydrodynamics in southeastern Australia imply localised dispersal of larvae. Journal of Experimental Marine Biology and Ecology (1988) 124:175–203.[CrossRef][Web of Science]
Moore J. D., Trueman E. R. Swimming of the scallop, Chlamys opercularis (L.). Journal of Experimental Marine Biology and Ecology (1971) 6:179–185.[Medline]
Nash R. D. M., Geffen A. J. The influence of nursery ground processes in the determination of year-class strength in juvenile plaice Pleuronectes platessa L. in Port Erin Bay, Irish Sea. Journal of Sea Research (2000) 44:101–110.[CrossRef]
Neill W. H., Miller J. M., van der Veer H. W., Winemiller K. O. Ecophysiology of marine fish recruitment—a conceptual framework for understanding interannual variability. Netherlands Journal of Sea Research (1994) 32:135–152.
Nell J. A., Holliday J. E. Effects of salinity on the growth and survival of Sydney rock oyster (Saccostrea commercialis) and Pacific oyster (Crassostrea gigas) larvae and spat. Aquaculture (1988) 68:39–44.[CrossRef][Web of Science]
Orensanz J. M., Parma A. M., Iribarne O. O. Population dynamics and management of natural stocks. Shumway S. E., ed. (1991) Amsterdam: Elsevier. 625–713. Scallops: Biology, Ecology and Aquaculture.
Peterson C. H. Recruitment overfishing in a bivalve mollusc fishery: hard clams (Mercenaria mercenaria) in North Carolina. Canadian Journal of Fisheries and Aquatic Sciences (2002) 59:96–104.
Ricker W. E. Stock and recruitment. Journal of the Fisheries Research Board of Canada (1954) 11:559–623.
Rothschild B. J. "Fish stocks and recruitment": the past thirty years. ICES Journal of Marine Science (2000) 57:191–201.
Seed R. Shell growth and form in the Bivalvia. Rhoads D. C., Lutz R. A., eds. (1980) New York: Plenum Press. 23–67. Skeletal Growth of Aquatic Organisms: Biological Records of Environmental Change, 1.
Shepard A. N., Auster P. J. Incidental (non-capture) damage to scallops caused by dragging on rock and sand substrates. Shumway S. E., Sandifer P. A., eds. (1991) Baton Rouge: World Aquaculture Society. 219–230. An International Compendium of Scallop Biology and Culture.
Shepherd J. G. A versatile new stock–recruitment relationship for fisheries, and the construction of sustainable yield curves. Journal du Conseil International pour l'Exploration de la Mer (1982) 40:67–75.
Sinclair M., Mohn R. K., Robert G., Roddick D. L. Considerations for the effective management of Atlantic scallops. Canadian Technical Report of Fisheries and Aquatic Sciences (1985) 113. 1382.
Sissenwine M. P. Why do fish populations vary? In: Exploitation of Marine Communities—May R. M., ed. (1984) Berlin: Springer. 59–95.
Smith P. E. Balancing sampling precision and fisheries management objectives—minimal methods. Bulletin of Marine Science (1993) 53:930–935.[Web of Science]
Sparre P., Ursin E., Venema S. C. Introduction to tropical fish stock assessment, 1. Manual. (1989) 337.
Stokesbury K. D. E., Himmelman J. H. Biological and physical variables associated with aggregations of the giant scallop Placopecten magellanicus. Canadian Journal of Fisheries and Aquatic Sciences (1995) 52:743–753.
Stoner A. W., Ray-Culp M. Evidence for Allee effects in an over-harvested marine gastropod: density-dependent mating and egg production. Marine Ecology Progress Series (2000) 202:297–302.[Web of Science]
Thouzeau G. Experimental collection of postlarvae of Pecten maximus (L.) and other benthic macrofaunal species in the Bay of Saint-Brieuc, France. 1. Settlement patterns and biotic interactions among the species collected. Journal of Experimental Marine Biology and Ecology (1991) 148:159–179.[CrossRef][Web of Science]
Tremblay M. J., Loder J. W., Werner F. E., Naimie C. E., Page F. H., Sinclair M. M. Drift of sea scallop larvae Placopecten magellanicus on Georges Bank—a model study of the roles of mean advection, larval behaviour and larval origin. (1994) 41:7–49. Deep-Sea Research Part II—Topical Studies in Oceanography.
Underwood A. J. Experiments in Ecology: their Logical Design and Interpretation using Analysis of Variance. (1997) Cambridge, UK: Cambridge University Press. 504.
Valencia A. H. Programa de cómputo para el procesamiento de datos en dinámica poblacional de peces. (1988) Mexico: Universidad Nacional Autónoma de México. MSc thesis.
Walters C. J., Martell S. J. D. Fisheries Ecology and Management. (2004) Princeton: Princeton University Press. 448.
Yonge C. M. The evolution of the swimming habit in the Lamellibranchia. (1936) 3:77–100. Mémoires du Musée Royal d'Histoire Naturelle de Belgique.
Young P. C., Martin R. B. The scallop fisheries of Australia and their management. CRC Critical Reviews in Aquatic Sciences (1989) 1:615–638.
Young P. C., McLoughlin R. J., Martin R. B. Scallop (Pecten fumatus) settlement in Bass Strait, Australia. Journal of Shellfish Research (1992) 11:315–323.
Zhang J. Z. Further studies of scallop spat settlement and early survival in Isle of Man coastal waters. (1996) UK: University of Liverpool. 254. PhD thesis.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||