© 2006 International Council for the Exploration of the Sea
Pollock (Pollachius virens) stock structure in the Canadian Maritimes inferred from mark-recapture studies
a Department of Fisheries and Oceans, Biological Station 531 Brandy Cove Road, St Andrews, NB, Canada E5B 2L9
b Department of Fisheries and Oceans, Bedford Institute of Oceanography Dartmouth, Nova Scotia, Canada B2Y 4A2
*Correspondence to J. D. Neilson: tel: +1 506 5298854; fax: +1 506 5295862. e-mail: neilsonj{at}mar.dfo-mpo.gc.ca.
The current management unit for pollock on the Canadian Atlantic coast is large compared with other gadoid resources, and includes the Scotian Shelf, the Bay of Fundy, and the Canadian portion of Georges Bank. Based on an analysis of mark-recapture studies conducted in the Canadian Maritimes and off southwestern Newfoundland and a review of other published studies providing data relevant to stock identification, the stock structure of pollock in Canadian Atlantic waters was re-assessed. The analysis also includes a novel method for using the spatial distribution of standardized fishing effort to predict the distribution of tag returns. It is concluded that three stocks co-occur within the current management unit. The larger population components exist in the western Scotian Shelf (including the eastern Bay of Fundy) and on the eastern Scotian Shelf. There is a coastal population in the western Gulf of Maine that overlaps into Canadian waters, but its size is likely to be relatively small. There is a need to revise the current management unit boundaries to protect the eastern Scotian Shelf stock, which on the basis of growth rate data, appears to be the least productive component of the pollock resource in Canadian Atlantic waters.
Keywords: mark-recapture, pollock, saithe, stock
Received 11 April 2005; accepted 20 December 2005.
 |
Introduction |
|---|
 Top Introduction MethodsResultsDiscussionReferences |
|---|
Pollock (Pollachius virens) is a commercially significant demersal fish species that occurs off the eastern coast of North America. For Canadian management purposes, the southernmost limit of the stock is considered to include fish in the Canadian portion of Georges Bank and the Gulf of Maine north to the Laurentian Channel (Figure 1). In comparison with other groundfish resources in the area, the Canadian management unit for pollock is large. Atlantic cod (Gadus morhua) and haddock (Melanogrammus aeglefinus), as examples, are each managed as three discrete units over the same geographic limits as the pollock resource. If the management unit for pollock does not reflect the stock complexity of the resource appropriately, there is potential for overexploitation of less productive components, or unrealized opportunities for harvesting more productive components. Stephenson (1999) further points out that failure to recognize or to account for complex stock structure could result in a loss of biological diversity, with unknown ecological consequences.
|
The initial management unit for the pollock resource was established by the International Commission for the Northwest Atlantic Fisheries (ICNAF, 1973), and included Division 4X (Figure 1) and Subarea 5 (includes the Gulf of Maine and Georges Bank area). At the time, it was noted that the stock structure was not well understood. In 1974, although no new information on stock structure was available, ICNAF considered it prudent to treat pollock on the Scotian Shelf and the Gulf of Maine as a single stock. Therefore, the management unit was extended to include Divisions 4V and 4W. After the establishment of the International Maritime Boundary in the Gulf of Maine area between Canada and USA in 1984, Canada reviewed the stock structure of marine resources in the area (Bowen, 1987), including information on meristics, morphometrics, genetics, and preliminary analyses of the available mark-recapture studies. It was concluded that the amount of transboundary movement was not sufficient to seriously affect stock conservation benefits if Canada undertook unilateral management actions within its own waters.
Compared with other groundfish resources off eastern North America, the coastal life history of juvenile pollock makes them uniquely amenable to tagging programmes. Clay et al. (1989) provided a synthesis of larval pollock length composition information from ichthyoplankton surveys, and from coastal mark-recapture studies of juvenile pollock conducted on a monthly basis over several years. Based on those observations, it seems that pollock spawn offshore, recruit to coastal environments as age-0 juveniles in midsummer, and remain there for about two years. Juvenile fish are readily caught in coastal gear such as traps, and individually by angling.
Among the tools available for study of stock structure, mark-recapture investigations are among the most powerful, especially when tags are applied at an early age (Ihssen et al., 1981). In this study, our null hypothesis is that regardless of point of release, tagged fish will distribute themselves uniformly throughout the current management unit. This is expected if the management unit is consistent with the biology of the exploited population, and if fish are moving freely throughout the range. Further, we predict that if the null hypothesis was true, the spatial distribution of recaptures within the management unit should be in proportion to the spatial distribution of effort. In addition to evaluating the adequacy of the current management unit for reflecting internal stock structure, we also describe the emigration and immigration of pollock from the current management area.
|
Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
For the purposes of this paper, we use the definition of Booke (1981) to define the term "stock": a population of fish that maintains itself over time in a particular area. Further, we consider the terms "stock" and "population" to be synonymous, so we use them interchangeably throughout the paper.
Over the period 1978–1984, an extensive pollock tagging programme was conducted in the Canadian Maritimes and off southwest Newfoundland (Clay et al., 1989). Tagging was done using dart tags with nylon T-anchors (Floy Manufacturing, Seattle, WA, USA). Tags were inserted in the dorso-lateral region, posterior to the first dorsal fin. Releases totalled 55 857 pollock. Fish marked and released were typically aged 2 and younger (Neilson et al., 2003). The sites of release and numbers of releases by year are provided in Figure 2, and details of the length range of released fish by date and release location are given in Table 1. Almost all the fish that were marked and released were caught in inshore traps or by handlining. This approach yielded fish that were in good condition. Previous attempts to capture larger fish in bottom trawl operations were unsuccessful, because the fish so captured were moribund even after short (10 min) duration sets.
|
|
Recaptures totalled 5088 individuals (about 9% of the total releases; Table 2). For the statistical analyses presented later in this paper, we include only those fish that had been at liberty for at least 30 days, and where Unit Area of recapture was known. This reduced the number of fish available for the analysis to 3650 (Table 2). Neilson et al. (2003) compared the increments of growth achieved by marked pollock released on the eastern (4VW, Figure 1) and western (4X) components of the management unit, and found that the growth rates of fish released in the western component were significantly higher than those from the eastern component. Those authors suggested that such differences reflected a more complex stock structure than implied by the current management unit. To evaluate if the mark-recapture information provided in this paper supports the inference of stock discreteness within the current management unit, we divided the recapture information into eastern (4VW) and western (4X) components.
|
Estimates of mixing and migration rates between putative stocks, and inferences concerning stock structure can be seriously compromised by different tag recovery and reporting rates. As noted by Schwarz and Arnason (1990), the number of tag returns depends upon the product of migration rate (or homing rate) and the tag-recovery rate. The tag-recovery rate is considered to be proportional to the harvest rate. Therefore, to make meaningful inferences concerning stock structure from mark-recapture information, the influence of the tag-recovery rate must be considered. As the pollock fishery employs a variety of gear types, of which only some have recorded effort, we analyse the data in two ways. Where such data exist, effort data are used to adjust the distribution of tag recapture information to reflect the distribution of effort. For other fisheries where no reliable information on fishing effort is available, the unadjusted tag-recapture data are presented.
The mobile gear (side and stern otter trawl) fishery accounted for 71% of the total pollock catch during the period that tag returns were obtained (1979–1990), and effort information is available for that fishery. As pollock are caught as bycatch in other groundfish fisheries as well as in directed pollock fisheries, we needed to describe the spatial distribution of all groundfish fishing effort. Detailed catch and effort data were available for mobile gear during the period of tag recaptures. The catch and effort data collected by the Canadian Department of Fisheries and Oceans provide species composition, date of landing, statistical area [which includes either Northwest Atlantic Fisheries Organization (NAFO) Statistical Divisions or at a finer geographic scale, Departmental statistical divisions referred to as Unit Areas (Figure 1)], gear and mesh type, and tonnage class, among other information.
We obtained standardized groundfish fishing effort for the period when tags were returned using a multiplicative model (Gavaris, 1980). The multiplicative model allowed us to examine the relationships between pollock catch rate and factors such as statistical area, tonnage class, month, groundfish target species, and trawler type (stern and side trawler), and to obtain estimates of standardized mobile gear effort for each Unit Area over the period when tags were returned. With knowledge of the distribution of standardized effort by Unit Area over the period that tags were returned, it is possible to develop an expected distribution of recaptures by Unit Area assuming that tagged fish move freely throughout the management unit. This was done by distributing the total number of recaptured fish among Unit Areas, pro-rated by the proportions of standardized effort by Unit Area from 1979 to 1990, the period of recaptures for the marked pollock. Observed distributions of recaptures by Unit Area were summed for the eastern and western components, then compared with the expected number of recaptures for the eastern and western components, using a 
2 statistical approach. Rejection of the null hypothesis meant that observed recapture distributions significantly differed from the expected distribution, and was considered evidence that fish were not mixing freely throughout the current management unit. Such a result would bring the validity of the current management unit into question.
For fisheries where no effort data were available or when the available recapture data were relatively sparse, the results are provided graphically.
Finally, we examined the spatial distribution of presumed spawning fish by selecting recaptures made during the period of spawning as determined from ichthyoplankton surveys (November through February inclusive; Neilson and Perley, 1996), and from fish whose length at recapture was >50 cm. Pollock of such length exceeded the length at first maturity reported by Trippel et al. (1997) for the Scotian Shelf.
|
Results |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Characteristics of released and recaptured fish
The distribution of days at liberty for recaptured fish is shown in Figure 3. The median days at liberty were 536 and 225 for the east and west recaptures, respectively, and significantly differed (Mann–Whitney U-test: U = 2 718 067, p < 0.001). We also examined the distribution of days at liberty for fish recaptured in fixed and mobile gear, and found that fish recaptured in mobile gear were at liberty for a significantly longer period (Mann–Whitney U-test: U = 345 537, p < 0.001). The distributions of lengths at release and recapture are shown in Figure 4. The average size at release of fish in the eastern portion of the management unit was significantly larger (independent samples t-test: t = 93.46, p < 0.001) than that of fish released from the west (means = 28.4 and 22.6 cm, for east and west, respectively). The average size at recapture was marginally different (independent samples t-test: t = –2.41, p = 0.016) between those released in the eastern and western components (means = 55.62 and 57.67 cm, for east and west, respectively). The distribution of month of recaptures is shown in Figure 5, and indicates a trend of increasing numbers of recaptures towards midyear. The pattern of recoveries tended to follow the seasonality in the landings of the fishery, as indicated in Figure 5. An exception was a large (n = 621) catch of tagged fish in December 1981, 33–39 days after release in the same trap that initially captured them.
|
|
|
Movements within the management unit
Results of the effort standardization for mobile gear during the period of tag recaptures are shown in Table 3. All factors (Unit Area, month, groundfish species targeted, and tonnage class) included in the analysis were significant (p < 0.05) in accounting for variation in pollock catch rates. The resulting distribution of standardized mobile gear fishing effort by Unit Area is reported in Table 3 and shown graphically by Unit Area in Figure 6. The standards chosen were Unit Area 4Vn, month = April, trawler type = stern, tonnage class = 4, and species directed for = pollock. Standardized effort was computed for each Unit Area using those standards. Areas of particular concentration of mobile gear effort associated with the pollock fishery included Unit Areas 4Vsc, 4Vn, 4Xn, and 5Zj. Using the distribution of standardized mobile gear effort among all Unit Areas comprising the management unit, and assuming that tagged fish moved freely throughout the management unit, we predicted the distribution of recaptured fish by the mobile gear fishery by Unit Area and compared them with the observed distribution (Figure 7).
|
|
|
Of the 1159 recaptures made by mobile gear from eastern releases, 72% (830) were made in the eastern side of the management unit. As expected given the distribution of fishing effort, the greatest number of recaptures was made in Unit Area 4Vsc (Figure 7). However, greater numbers of recaptures than expected were made in Unit Areas 4Wh and 4Wk. Considering the movement of eastern-released fish into the western area, observed recaptures were close to the expected numbers only in the easternmost portion of the western area (4Xm and 4Xn; Figure 6). Aggregating all recapture information from eastern releases into either eastern recaptures or western recaptures and comparing them against expected recaptures given the standardized effort distribution presented in Figure 6, we rejected the null hypothesis that eastern-released pollock are recaptured in proportion to the distribution of recapture effort (
2 = 140.6; p < 0.001).
Of the 501 recaptures made by mobile gear from western releases, 81% of recaptured fish were caught within the western side of the management unit. The largest number of recaptures was from Unit Area 4Xq and greatly exceeded the expected number (158 and 38.2, for observed and expected recaptures, respectively). Aggregating all recapture information from western releases into either eastern recaptures or western recaptures and comparing them against expected recaptures given the standardized effort distribution presented in Figure 6, we rejected the null hypothesis that western released pollock are recaptured in proportion to the distribution of recapture effort (2 = 418.8; p < 0.001).
In general, the spatial distribution of recaptured fish in other fisheries was similar to that observed in the mobile gear fishery (Figure 8), and did not indicate a tendency to mix freely throughout the management unit. Area 4Wd, however, was associated with a considerably higher (70%) proportion of recaptures in the fixed-gear fishery than in the mobile gear fishery (1%). Area 4Wd also had the greatest number (486) of recaptures of eastern-released fish of all Unit Areas. Of the pollock released on the eastern side of the management unit, 83% were recaptured there.
|
From the western releases recaptured in other fisheries, the largest number of recaptures was noted in 4Xo (1066; Figure 8). Of the pollock released on the western side of the management unit, more than 98% were recaptured there, only four recaptures coming from the eastern side.
Movement to areas outside the management unit
Of the 3632 recaptured fish included in this study that originated from releases within the current management unit, 3516 (97%) were recaptured within the boundaries of the management unit. Of the fish that moved outside the management unit, 22 were recaptured in waters around Newfoundland (includes NAFO Subareas 2 and 3), 5 were recaptured in the northern Gulf of St Lawrence, and 88 were recaptured in USA waters. Of the 22 recoveries in Newfoundland waters, 14 were from the nearest point of release adjacent to the northern limit of the management unit (4Wd, Figure 9). All recaptures in the Gulf of St Lawrence also originated from that point of release. Of the recaptures in waters around Newfoundland, 8 of the 22 were reported as being caught off northern Newfoundland and Labrador (NAFO Division 3K). Given that the tonnage of pollock landed in that area was extremely small during the period of recaptures, we consider that these reports to be of uncertain validity, and consequently the returns from northern Newfoundland and Labrador are not shown in Figure 9.
|
Releases from five sites crossed into USA waters, with 63 originating from the western Bay of Fundy, 10 from the eastern Bay of Fundy, 5 from southwest Nova Scotia, 1 from the Halifax area, and 9 from the easternmost release site. Some of the longest distances between release and recapture were noted for the latter recaptures, with movements of >400 nautical miles noted for individual fish.
The difference in the number of recaptures moving into USA waters from the releases from the eastern and western sides of the Bay of Fundy is noteworthy (Figure 10). From the releases of pollock on the western side of the Bay of Fundy (n = 15 717), there were 403 recaptures, of which 63 were taken in USA waters (15.8% of all recaptures). From the eastern side, there were 6081 releases and 255 recaptures, of which 10 were captured in USA waters (3.9% of all recaptures).
|
Movements into the management unit
Of 312 fish released off southern and western Newfoundland, 33 were recaptured, and recapture coordinates were reported in 18 cases (Figure 11). In all, 17 were recaptured off southern Newfoundland, 8 in the Gulf of St Lawrence, and the remaining 8 on the Scotian Shelf. No mark-recapture experiments involving pollock have been undertaken in USA waters, so it is not possible to comment on the potential immigration across the southern boundary of the management unit.
|
Evidence for spatial segregation of spawning components
The location of presumed spawning fish (>50 cm, captured during the spawning period of November through February inclusive) is shown in Figure 12. We show the distribution of recaptures of presumed spawners originating from release sites at the western (Area 4Xs, Table 1) and eastern (Area 4Wd) extremities of the management unit. While there is some overlap in the distribution of the presumed spawners, recaptures from the western release predominate in the Gulf of Maine, and only rarely on the eastern Scotian Shelf. Recaptures from the eastern release site occur throughout the Scotian Shelf, but only rarely in the Gulf of Maine, with the exception of the northeast corner of Georges Bank.
|
|
Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
As noted by Gulland (1983), inferences regarding stock structure can be assessed using five types of information. He considered that a discontinuous geographic distribution of fishing (i) could suggest a gap in the distribution of fish, which may correspond to a separation of stocks. Such information could also be obtained from fishery-independent means, such as research vessel surveys. Information on spawning areas (ii) is critical, because genetic separation of stocks requires a clear separation of spawning components, even if such fish mix at a later stage. Gulland (1983) considered that values of population parameters (iii) may provide insight if there are stock differences, citing parameters such as growth rate and mortality as examples. He noted that morphological or physiological characteristics that are genetically determined (iv) can provide clear evidence that two groups are distinct, but genetic separation can, in theory, exist without being evident in the characteristics examined. Lastly, Gulland (1983) notes that tagging (v) can, in principle, provide the clearest evidence of stock separation or otherwise. We next evaluate the available data in Gulland's (1983) five categories of information that are useful for the assessment of pollock stock structure.
Neilson et al. (2002) provide a summary of the distribution of catches from both research vessel surveys and the commercial fishery. The monthly distribution of catches by large otter trawlers was patchy, and concentrated in NAFO Division 4X, with additional fishing activity closely associated with the 200-m contour throughout the rest of the management unit. During the period of spawning from 1981 to 2000 (November–February, based on examination of ichthyoplankton records; Neilson and Perley, 1996), there was relatively little fishing activity in NAFO Division 4W, with evidence of spatial discontinuity in the fishery in that area. Research vessel surveys conducted in March between 1979 and 1984 (Neilson et al., 2002) also show a spatial discontinuity in the eastern portion of NAFO Division 4W.
Considering next the available information on the distribution of spawning activity, Hanke et al. (2001) provided the most complete information on the distribution of pollock eggs and larvae within the current management unit. When they evaluated the composite distribution of egg stages and larvae from the available ichthyoplankton surveys conducted from 1975 to 1997, discrete aggregations of early life history stages were found on Georges Bank, off southwestern Nova Scotia (NAFO Division 4X), and in Divisions 4VW. Clear spatial discontinuities existed among the aggregations described above. Cargnelli et al. (1999) summarize available (1978–1987) USA information on ichthyoplankton distribution in the 5Y, 5Z, and 4X areas, and as such, these data provide insight into stock separation at the southern limit of the Canadian management unit. The USA surveys included a north–south transect that had very closely spaced stations through the Fundian Channel. While that bathymetric feature separated the aggregations apparent in the Canadian ichthyoplankton surveys described earlier, the USA surveys often caught eggs at almost all the stations along the transect. Therefore, the apparent division of the Georges Bank and NAFO Division 4X aggregations described earlier may be an artefact of the distribution of sampling. Further west and closer inshore, occurrences of eggs were sometimes found in USA surveys that extended to the northeast along the New England coast, although the sampling density in that area was low compared with the rest of survey area. Considering the distribution of presumed spawners, the data presented here indicate that fish tagged in the western portion of the management unit tend to be recaptured as presumed spawners in the Gulf of Maine most frequently. Fish tagged in the eastern portion of the management unit tend to be recaptured as presumed spawners most frequently on the Scotian Shelf. Taken together, the ichthyoplankton and distribution of presumed spawners indicate the occurrence of multiple spawner areas throughout the management unit.
Regarding population parameters, Neilson et al. (2003) examined the increment of growth achieved by fish released in the eastern (4VW) and western (4X) sides of the current pollock management unit. We have here illustrated that the growth rate of pollock released in the western side of the management unit was significantly faster than those released in the east. These results may reflect either genetic differences or a more favourable environment for growth. If the former explanation is valid, a more complex stock structure than the current management unit is implied. Trippel et al. (1997) also found that the age and length at first maturity for pollock on the eastern Scotian Shelf were greater than the corresponding values for the western Scotian Shelf.
McGlade (1983) presented meristic information on pollock collected during research surveys conducted in the Scotian Shelf – Georges Bank and Gulf of Maine region. She concluded that based on multivariate analyses, there were clear divisions among adult fish collected in USA waters in the western Gulf of Maine, Browns and Roseway Banks (4X), and Emerald and Western Banks (4VW). McGlade and Boulding (1987) employed a morphometric approach called truss analysis to confirm the separation of adult pollock collected on the Scotian Shelf and the western Gulf of Maine. In contrast, Mayo et al. (1989) noted that electrophoretic analysis of 28 enzyme systems in pollock tissue from Emerald Basin (Scotian Shelf) and Jeffreys Ledge (western Gulf of Maine) revealed no significant differences. Cargnelli et al. (1999) suggest that the genetic differentiation of pollock off eastern North America could be better resolved using more advanced genetic (e.g. microsatellite DNA markers) and biochemical techniques (e.g. elemental fingerprinting of otoliths).
The present study provides insight into stock structure from data obtained from mark-recapture studies, the fifth and final category of Gulland's (1983) list of information that is of consequence to stock identification studies. The tagging data indicate that fish do not move freely within the management unit. Of those fish marked and released in the western side of the management unit, a high proportion (98%) was recaptured there. Fish marked and released in the east showed a greater propensity to move significant distances, with some recaptures on Georges Bank. Overall, however, 72% were recaptured in the eastern side of the management unit.
Taken together, the available published evidence and the new data presented here provide a compelling view that the current management unit does not adequately reflect the complexity of stock structure in the Canadian Maritimes. We believe that the tagging, growth rate, ichthyoplankton, fisheries distribution, and meristic and morphometric data provide a comprehensive basis to state that the current management unit should be subdivided into eastern and western components. Of the data available, only the electrophoretic study referred to by Mayo et al. (1989) provides evidence to the contrary. As pointed out by Begg and Waldman (1999), even when results of various studies of stock structure are not completely consonant, the default management scenario should be to use a precautionary approach to ensure resource sustainability and maintenance of genetic biodiversity. The precise location of the boundary between the eastern and western components is less clear, as Unit Areas 4Xm and 4Xn appear to be an area of overlap between the two components.
Having described movements within the management unit and the likely population structure, it is of obvious importance also to consider mixing with adjacent areas outside the current management unit. Considering first the linkages with USA waters, our results showed a significant (n = 63) movement of pollock marked and released off the western Bay of Fundy south into USA waters, with a notable concentration off the coast of Massachusetts (Jeffrey's Ledge, Stellwagen Bank area). Given that Mayo et al. (1989) indicated that there were relatively low levels of fishing effort in the western Gulf of Maine compared with Canadian waters during the period when tagged fish were at liberty, the returns from USA waters have added significance. Fish tagged and released off the eastern side of the Bay of Fundy did not show such a clear affinity with USA waters. Steele (1963), in describing the results of an earlier study of marked and released pollock off the western Bay of Fundy, also noted the recaptures of marked pollock off the Massachusetts coast during the period November–February. He suggested that the pollock population in the western Gulf of Maine followed a north–south seasonal migration, invading northern waters during early summer, then returning to the southern Gulf of Maine to reproduce. While our data support this interpretation in part, it is also clear that a significant proportion of the pollock marked in the western Bay of Fundy remain in Canadian waters. Therefore, we conclude that there is a third population of pollock in the current Canadian management unit. However, it is likely that this population is small in comparison with the two described earlier.
Mixing outside the management unit at the northeast boundary appears to be less widespread than the associations with USA waters at the southwestern boundary. Unlike the situation in the south where marked fish from five different points of release crossed the International Maritime Boundary, fish that crossed the northeastern boundary typically originated from the nearest point of release, with only two exceptions. From the limited releases outside the management unit off southwestern Newfoundland, 24% (8 of 33 recaptures) were recaptured in the current management unit. Although the number of released and recaptured fish was comparatively small, these data indicate a significant linkage between a juvenile nursery area off southwestern Newfoundland and the current management unit.
Overall, some 22% of fish marked in the eastern side of the management unit move into the western side, but only 2% of fish marked on the western side move to the east. The question arises regarding what level of mixing between adjacent stocks could be tolerated before the utility of the management unit is compromised. Kimura et al. (1998) used mark-recapture studies to elucidate the stock structure of sablefish (Anoplopoma fimbria) in northeastern Pacific waters. Of relevance to our study, Kimura et al. (1998) were interested in the hypothesis that sablefish off Alaska and Canada north of 50°N form a separate population from those to the south. They noted that over the course of their study (1971–1993), approximately 3.5% of fish marked in the north moved to the south, and approximately 4.4% of fish marked in the south moved north. They observed that annual rates would be considerably lower, and concluded that the sablefish populations examined form separate populations that, for practical management considerations, remain largely independent.
In the Northeast Atlantic, pollock (also called "saithe") form a significant part of the demersal fish community, and are the basis of a significant commercial fishery. The saithe resource is considered to consist of five discrete stocks, including the Northeast Arctic, North Sea, Iceland, Faroe Islands, and West of Scotland. Jakobsen and Olsen (1987) reported the results of an extensive programme of mark-and-release work involving immature fish tagged off northern Norway between 1954 and 1980. After excluding recaptures that were judged to be too soon after release, 549 recaptures were made. Of these, 106 (19%) were taken in the Iceland management unit and 55 (10%) in the Faroes Island management unit. Such migration rates are comparable with those observed in our study, yet a complex management scheme involving five management units is maintained.
Hunt et al. (1999) examined movement of Atlantic cod (Gadus morhua) in the Gulf of Maine area. They noted an exchange of about 15% between NAFO Divisions 4X and 5Z, with somewhat higher rates between 4X and 5Y. Once again, these mixing rates are roughly comparable with those observed in our study, and the USA and Canada maintain four management units over this area. Therefore, from these three examples from the published literature, it appears as though discrete management units for gadoid stocks are maintained in the face of mixing rates that are equal to or exceed those observed in this study.
Our study has attempted to adjust the recapture data for the observed spatial distribution of standardized effort by the mobile gear fishery, and represents one of only few such examples in the literature pertaining to mark-recapture studies of fish populations. However, as noted by Kimura et al. (1998), if the distribution of standardized fishing effort is mis-specified, the resulting inferences may be further distorted rather than corrected. In this regard, it is significant that the inferences concerning stock structure presented here use data from two sources, including standardized (mobile gear) and unstandardized (other fishing gear). Results from the two sources gave similar indications of mixing between the two sides of the management unit, but provide sometimes quite different impressions of the relative significance of a given Unit Area when comparing the spatial distribution of tag returns. For example, just 10 fish were recaptured by mobile gear in Unit Area 4Wd, an inshore area close to a major release site, compared with 475 recaptures from the fixed-gear fishery. Such discrepancies reflect the differential distribution of fishing effort: there was very little mobile gear effort in Unit Area 4Wd compared with the fixed-gear fishery. To some extent, the limited movement of fish tagged in the western part of the management unit to the eastern part and vice versa could be attributed to the location of the main sites of release close to opposite extremes of the management unit. Nonetheless, our results clearly indicate that pollock do not move freely within the current management unit.
Our results have shown that the time at liberty for those pollock recaptured by fixed gear was significantly less than time at liberty for those recaptured by mobile gear. It may be that the minimum days at liberty that we accepted (
30 d) in this study was too short for this gear type, given the largely inshore distribution of fixed-gear fishing effort. If that were true, tagged fish may have had insufficient time to disperse. Given this observation, and the lack of opportunity to adjust for the spatial distribution of fishing effort for this gear type, we suggest that the results from the mobile gear analyses carry more weight.
The entire management unit does not provide preferred habitat for pollock, and it may be that the ratio of observed returns to expected returns given the distribution of standardized mobile gear effort provides an index of habitat suitability in each Unit Area. Values <1 reflect a lower number of recaptures than expected if the marked fish were mixing freely throughout the management unit. Conversely, values >1 would reflect habitat that is preferred by pollock. As examples, Unit Area 4Xq has considerably more observed recaptures than expected from western releases (observed 158, expected 38.2), as did Unit Area 4Wk (observed 150, expected 27.7) from eastern releases. Habitat preferred by pollock includes areas of contrasting bottom depth (Neilson et al., 2002). In the case of 4Wk, the Unit Area includes the edge of the Scotian Shelf and 4Xq includes areas of mixed depth at the mouth of the Bay of Fundy (Figure 1). Of course, inferences of habitat quality based on the observed and expected returns of tags would be confounded by management measures such as closed areas or seasons that influence the distribution of fishing effort.
|
Acknowledgements |
|---|
We are grateful to all those within our Division who conducted the tagging work between 1978 and 1985, and to Mark Fowler for his assistance in assembling the tagging data, and for providing details of the methods. We also thank Erin Carruthers, Ralph Mayo, and an anonymous reviewer for their constructive reviews of an early draft of this paper.
|
References |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
-
Begg G.A. and Waldman J.R. (1999) An holistic approach to fish stock identification. Fisheries Research 43:35–44.[CrossRef][Web of Science]
Booke H.E. (1981) The conundrum of the stock concept – are nature and nurture definable in fisheries science? Canadian Journal of Fisheries and Aquatic Sciences 38:1479–1480.
Canadian Atlantic Fisheries Scientific Advisory Committee Research Document Bowen D. B. (1987) A review of stock structure in the Gulf of Maine Area: a workshop report. 87/21: 51 pp.
Cargnelli L. M., Griesbach S. J., Packer D. B., Berrien P. L., Johnson D. L., Morse W. W. (1999) Pollock, Pollachius virens, life history and habitat characteristics. NOAA Technical Memorandum, NMFS-NE-131. 30 pp.
Clay D., Stobo W.T., Beck B., Hurley P.C.F. (1989) Growth of juvenile pollock (Pollachius virens L.) along the Atlantic coast with inferences of inshore–offshore movements. Journal of Northwest Atlantic Fisheries Science 9:37–43.
Gavaris S. (1980) Use of a multiplicative model to estimate catch rate and effort from commercial data. Canadian Journal of Fisheries and Aquatic Sciences 37:2272–2275.
Gulland J. A. (1983) Fish Stock Assessment. FAO/Wiley Series on Food and Agriculture, vol. 1. 223 pp.
Hanke A. R., Page F. H., Neilson J. D. (2001) Distribution of pollock (Pollachius virens) eggs and larvae on the Scotian Shelf, eastern Gulf of Maine, Bay of Fundy and eastern Georges Bank. Canadian Technical Report of Fisheries and Aquatic Sciences, 2345. 118 pp.
Hunt J.J., Stobo W.T., Almeida F. (1999) Movement of Atlantic cod, Gadus morhua, tagged in the Gulf of Maine area. Fishery Bulletin US 97:842–860.
Ihssen P.E., Booke H.E., Casselman J.M., McGlade J.M., Payne N.R., Utter F.M. (1981) Stock identification: materials and methods. Canadian Journal of Fisheries and Aquatic Sciences 38:1838–1855.
International Commission for the Northwest Atlantic Fisheries (ICNAF). (1973) Special Commission Meeting – January 1973. Proceedings No. 3, Appendix VI. p. 22.
Jakobsen T. and Olsen S. (1987) Variation in rates of migration of saithe from Norwegian waters to Iceland and Faroe Islands. Fisheries Research 5:217–222.[CrossRef][Web of Science]
Kimura D.K., Shimada A.M., Shaw F.R. (1998) Stock structure and movement of tagged sablefish, Anoplopoma fimbria, in offshore northeast Pacific waters and the effects of El Niño – Southern Oscillation on migration and growth. Fishery Bulletin US 96:462–481.
Mayo R.K., McGlade J.M., Clark S. (1989) Patterns of exploitation and biological status of pollock (Pollachius virens L.) in the Scotian Shelf, Georges Bank, and Gulf of Maine Area. Journal of Northwest Atlantic Fishery Science 9:13–36.
McGlade J. M. (1983) Preliminary study of the stock structure of pollock (Pollachius virens L.). Canadian Atlantic Fisheries Scientific Advisory Committee, 83/81. 21 pp.
Canadian Technical Report of Fisheries and Aquatic Sciences McGlade J. M. and Boulding E. G. (1987) The truss: a geometric and statistical approach to the analysis of form in fishes. 1457: 34 pp.
Neilson J.D., Annis L., Perley P., Clay A., Croft C., O'Connor M. (2002) Seasonal aggregations of Canadian East Coast pollock (Pollachius virens (L.)), as inferred from the commercial fishery and hydroacoustic observations. Journal of Fish Biology 61:1067–1084.[CrossRef][Web of Science]
Neilson J.D. and Perley P. (1996) In Burke D.L., O'Boyle R.N., Partington P., Sinclair M. (Eds.). Can ichthyoplankton data be used to describe spawning areas of marine fish? Report of the Second Workshop on Scotia-Fundy Groundfish Management pp. 129–146 Canadian Technical Report of Fisheries and Aquatic Sciences, 2100.
Neilson J.D., Stobo W.T., Perley P. (2003) Age and growth of Canadian East Coast pollock: comparison of results from otolith examination and mark-recapture studies. Transactions of the American Fisheries Society 132:536–545.[CrossRef]
Schwarz C.J. and Arnason A.N. (1990) The use of tag-recovery information in migration and movement studies. American Fisheries Society Symposium 7:588–603.
Steele D.H. (1963) Pollock (Pollachius virens (L.)) in the Bay of Fundy. Journal of the Fisheries Research Board of Canada 20:1267–1314.[Web of Science]
Stephenson R.L. (1999) Stock complexity in fisheries management: a perspective of emerging issues related to population sub-units. Fisheries Research 43:247–249.[CrossRef][Web of Science]
Canadian Technical Report of Fisheries and Aquatic Sciences Trippel E. A., Morgan M. J., Frechet A., Rollet C., Sinclair A., Annand C., Beanlands D., Brown L. (1997) Changes in age and length at sexual maturity of Northwest Atlantic cod, haddock and pollock stocks, 1972–1995. 2157: 120 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Armannsson, S. Th. Jonsson, J. D. Neilson, and G. Marteinsdottir Distribution and migration of saithe (Pollachius virens) around Iceland inferred from mark-recapture studies ICES J. Mar. Sci., July 1, 2007; 64(5): 1006 - 1016. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||