Estimation of sea scallop abundance using a video survey in off-shore US waters.

ABSTRACT A video survey was conducted from May 28, 2003, to August 23, 2003, to provide spatially explicit estimates of sea scallop density and size distributions along the off-shore northeast waters of the United States. Sea scallop (Placopecten magellanicus) densities in the Mid-Atlantic (26,270 [km.sup.2]) and Georges Bank (28,523 [km.sup.2]) ranged from 0.04 to 0.79 and 0.09to 0.26 scallop * [m.sup.-2] respectively, and represented approximately 217,520 mt tons of scallop meats (approximately US$2.4 billion). Sea scallops were highly aggregated in areas closed to mobile fishing gear. In the Georges Bank closed areas the proportion of sea scallop pre-recruits(<90 mm shell height) was low and sufficient to replace the adult population at an instantaneous mortality rate of 0.10 but not at a higher rate. A large number of prerecruit scallops were observed in the southern portion of the Hudson Canyon closed area extending south into open waters. Sea stars outnumbered sea scallops (approximately 39 to 16 billion, respectively) although most were small (20- to 40-mm arm length). Sea stars may be responsible for sea scallop mortality in the southern portion of Closed Area II.

KEY WORDS: sea scallop, Placopecten magellanicus, video, systematic survey, sea stars, Georges Bank, mid-Atlantic

INTRODUCTION

Large densities of sea scallops, Placopecten magellanicus (Gmelin 1791), exist in areas of the Mid-Atlantic and Georges Bank that have been closed to commercial scallop dredging since 1998 and 1994, respectively. Three areas on Georges Bank were closed to all mobile fishing gear to protect declining groundfish stocks, and two areas in the Mid-Atlantic were closed to enhance scallop abundance (Murawski el al. 2000, Hart 2001, 2003). To use the sea scallop resource in these areas, the New England Fisheries Management Council and National Marine Fisheries Service (NMFS) are developing fisheries management plans that require spatially explicit information.

Estimates of sea scallop abundance are made using dredge surveys, which require an estimate of dredge efficiency to determine absolute densities (Caddy 1989, Murawski et al. 2000). Using a video survey eliminates the error associated with the dredge efficiency as the actual number of scallops within the sample area is counted (Stokesbury 2002). Furthermore, this system provides spatially explicit information on the sea floor and marine benthic community (Stokesbury 2002).

The objective of this study was to provide spatially explicit, accurate, precise, absolute estimates of sea scallop density and size distributions along the off-shore northeast waters of the United States. Detailed information on the sediment and marine benthic habitat associated with the sea scallop fishing grounds will be presented in a future publication.

MATERIALS AND METHODS

In preparation for the survey Letters of Authorization for scientific research were obtained from the NMFS for each participating sea scallop-fishing vessel. Commercial fishing gear was removed from the fishing vessels prior to conducting each six-day cruise. A mobile studio, including monitors and S-VHS video recorders for each camera, a monitor for the Captain controlling the vessel's hydraulic winches to deploy the pyramid, a laptop computer with Arcpad GIS[R] software integrated with a differential global positioning system with a WAAS receiver, and a laptop computer for data entry, was assembled in the wheelhouse. The survey grid was plotted prior to the cruise in Arcpad GIS[R]. Two scientists, a captain, mate and one deck hand were able to survey stations continually, completing 20 to 30 stations every 24 h.

The SMAST sampling pyramid was deployed from the scallop fishing vessels (Fig. 1). Two cameras mounted vertically on the pyramid at a height of 700 and 1575 mm above the pyramid's base provided quadrat sizes of 0.595 and 2.841 [m.sup.2], respectively (Figs. 2 and 3). A third camera mounted horizontally 50 mm above the pyramid base provided a side view across the sample area (Fig. 4). All sea scallops were counted, including those along the edge of the quadrat image that were only partially visible. To correct for this edge effect 75 ram, based on the average shell height of the scallops observed, was added to each edge of the quadrat image providing quadrat sizes of 0.788 and 3.235 [m.sup.2] (Stokesbury 2002).

[FIGURES 1-4 OMITTED]

Our primary quadrat sample size was 3.235 [m.sup.2], but one scallop per quadrat (0.31 scallop * [m.sup.-2]) is above minimal commercial densities (Caddy 1989, Brand 1991, Thouzeau et al. 1991, Stokesbury & Himmelman 1993, Stokesbury 2002). Therefore, at each station we collected four quadrat samples increasing the sample area to 12.94 [m.sup.2]. The vessel stopped at each station and the pyramid was lowered to the sea floor. Footage of the first quadrat was recorded, and then the pyramid was raised so the sea floor could no longer be seen. The vessel drifted approximately 50 m and then the pyramid was lowered to the sea floor again to obtain a second quadrat; this was repeated four times.

We used a centric systematic design to position the stations because it is simple, samples evenly across the entire survey area, and has been successfully used to survey scallops on Georges Bank (Thouzeau et al. 1991, Stokesbury 2002). With this sampling design it is possible to estimate densities of macroinvertebrates within different areas without violating the sampling protocol or paying a statistical penalty for post-stratification. Further the centric systematic design facilitates mapping sea floor sediments and macroinvertebrate distributions.

Stations were positioned on a 5.6 by 5.6 km (3.0 by 3.0 nautical mile) grid overlying historical and present fishing grounds based on information from commercial fishermen and the Vessel Monitoring System (Rago et al. 2000; Fig. 5). We selected the distance of 5.6 km between stations because it was logistically feasible and allowed an accurate estimate of the mean sea scallop density with little loss of precision. Estimates of mean sea scallop * [m.sup.-2] from the Nantucket Lightship area, surveyed in 2002, were similar for distances between stations ranging from 1.6 to 5.6 km; the standard error (SE) increased because of reduced sample size but the coefficients of variation (CV) were still low, 0.62 (SE = 0.057, CV = 9.3%) and 0.62 (SE = 0.101, CV - 16.3%) scallop * [m.sup.-2], respectively. For distances greater than 5.6 km between stations the estimated means of scallop * [m.sup.-2] as well as the standard errors and coefficients of variation increased, for example the mean was 0.75 scallop * [m.sup.-2] (SE = 0.300, CV = 40.1%) for a 9.3-km station grid.

[FIGURE 5 OMITTED]

Video footage of the sea floor was recorded on S-VHS tapes. For each quadrat the time, depth, number of live and dead scallops, and latitude and longitude were recorded (Fig. 6). After each survey the videotapes were reviewed in the laboratory and an image of each quadrat was digitized and saved using Image Pro Plus[R] software (TIF file format). Within each quadrat marcoinvertebrates and fish were counted and the substrate was identified (Stokesbury 2002). Counts were standardized to individuals * [m.sup.-2].

[FIGURE 6 OMITTED]

Sea scallop shell heights (mm) were measured using Image Pro Plus[R] software (Fig. 7). An equation (Equation 1) to correct for the curve of the camera lens was applied to each shell height measurement. This equation was estimated and verified with laboratory and field experiments that indicated a measuring error of less than 5% (author's unpublished data).

[FIGURE 7 OMITTED]

(1) c = sh [square root of ([(x).sup.2] + [(y).sup.2])/[(y).sup.2])]

where:

c = corrected shell height (mm)

sh = shell height (mmm) measured using image pro

y = vertical camera height from the base of the sampling pyramid

x = distance from the center of the quadrat (mm)

Mean densities and standard errors of scallops were calculated using equations for a two-stage sampling design (Cochran 1977).

The mean of the total sample is:

(2) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where:

n = primary sample units (stations)

[bar.x.sub.i] = sample mean per element (quadrat) in primary unit i (stations)

x = the mean over the two-stages.

The standard error of this mean is:

(3) S.E.([bar.x]) = [square root of (1/n ([s.sup.2]))]

where:

[s.sup.2] = [[summation of].sup.n] [([bar.x.sub.i] - x).sup.2]/(n - 1) = variance among primary unit (stations) means.

Because the sampling fractions were small, hundreds of scallops sampled compared with millions of scallops in the area, the finite population corrections were omitted simplifying the estimation of the standard error (Cochran 1977).

To examine the effect of variation between elements in the two-stage sampling design, sampling four quadrats per station, we created a simple formula in Excel (Microsoft, Seattle, WA) that randomly selected one of the four quadrats from each station and then calculated the means and variances for each survey area. This removed the subsampling component of the two-stage sampling design reducing it to an unaligned systematic sample, which is often superior to stratified random sampling (Cochran 1977). Using the Excel formula we generated 240 estimates of the means and determined the standard error (Glantz 2002).

The absolute number of scallops within a survey area was calculated by multiplying the mean number of scallops * [m.sup.-2] by the total area surveyed (Stokesbury 2002). Distributions of scallops * [m.sup.-2] were plotted using ArcGIS[R] software. Estimates of scallop meat weight (w) were derived from shell height (sh) frequencies for each area and length/weight regressions for each area to the nearest season and time of the survey. These equations were calculated from live dissections of sea scallops collected during fishing trips from open areas and from vessels that participated in the Sea Scallop Exemption Fishery in each of the closed areas (Table 2; Nantucket Lightship Area w = 1.21 x [10.sup.-5] * [sh.sup.3.1062], CAI w = 8.21 x [10.sup.-6] * [sh.sup.3.1321], CAII w = 1.52 x [10.sup.-5] * [sh.sup.3.0443], Georges Bank open w = 7.38 x [10.sup.-6] * [sh.sup.3.1559], mid-Atlantic open, HC and VA w = 1.63 x [10.sup.-5] * [sh.sup.2.9485; author's unpublished data).

RESULTS

The 2003 survey of commercial off-shore sea scallop fishing grounds covered 54,793 [km.sup.2] (16,000 n[m.sup.2]). Sea stars (Asteroidea) were the most abundant macroinvertebrate observed numbering approximately 39 billion, although most individuals were small (20-40 mm arm length) (Table 1). High densities of sea stars occurred in both open and closed areas of Georges Bank and the mid-Atlantic with the highest density in the Nantucket Lightship area (Table 1).

Sea scallops, Placopecten magellanicus, were the second most abundant macroinvertebrate observed numbering approximately 16 billion scallops representing approximately 217,520 mt Ions of scallop meats (Table 2).

The mid-Atlantic contained the greatest number of sea scallops observed in the 2003 survey, 74% (Table 2, Fig. 8). In the mid-Atlantic 23% of the area surveyed is closed to scallop fishing. The lowest and highest densities of scallops in the mid-Atlantic were in the Virginia Beach and Hudson Canyon closed areas, 0.04 and 0.79 scallop * [m.sup.-2], respectively (Table 2, Fig. 9). Assuming that stations with at least one scallop observed in four quadrats represents scallop habitat (Stokesbury 2002) the densities increased to 0.09 (SE = 0.007) to 1.08 (SE = 0.338) scallop * [m.sup.-2]. Sea scallops were aggregated in the Hudson Canyon closed area, which contained 33% of the scallops observed in the mid-Atlantic, representing 36% of the scallop meat weight, (comparison of the number of scallops observed in open and closed areas, df = 1, [chi square] - 290, P < 0.001).

[FIGURES 8-9 OMITTED]

The mid-Atlantic contained only 46% of the total scallop resource meat weight, (Georges Bank and mid-Atlantic combined; Table 2). This low meat weight compared with the high density of scallops is due to a large recruitment of scallops that were below commercially harvestable size (90-mm shell height equal to a 3.5" ring mesh size in the scallop dredge; Fig. 9). In the Hudson Canyon closed area and mid-Atlantic open area 62% and 69% of the scallops measured were less than 90 mm (Fig. 9).

Sea scallops were highly aggregated on Georges Bank with 66% of the all the individuals observed, representing 82% of the scallop meat weight, within the closed areas (comparison of the number of scallops observed in open and closed areas, df = 1, [chi square] = 330, P < 0.001; Table 2, Fig. 10). Densities of sea scallops ranged between 0.16 and 0.26 scallop, [m.sup.-2] in the closed areas of Georges Bank compared with 0.09 scallop * [m.sup.-2] in the open area (Table 2). Sea scallop densities increased to 0.38 (SE = 0.071) to 0.62 (SE = 0.094) scallop * [m.sup.2] in the closed areas and 0.23 (SE = 0.0175) scallop * [m.sup.-2] in the open area for stations where at least one scallop was observed within four quadrats.

[FIGURE 10 OMITTED]

The majority of scallops in the open area of Georges Bank, 59%, were below harvestable size (<90 mm shell height; Fig. 9). The majority of sea scallops in the closed areas were very large with 77%, 87%, and 87% of the scallops greater than 90-mm shell height in Closed Area I, Closed Area II, and the Nantucket Lightship Area, respectively (Fig. 9).

The two-stage sampling technique provided the same means as the unaligned systematic sample (Table 2). In all areas the CV decreased using the unaligned systematic sample except for the Virginia Beach closed area where both the number of stations and number of scallops observed were small (Table 2).

Scallops that had died of natural causes were identifiable in the video footage as the two shells were still attached at the umbo. The highest number of dead sea scallops occurred in the southern portion of Closed Area II (Fig. 11). A dense aggregation of large sea stars (9.20 sea star * [m.sup.-2], mean arm length 123.5 mm, SD = 35.39) was also observed in this area with the dead scallops suggesting a localized high natural mortality event (Fig. 12).

[FIGURES 11-12 OMITTED]

DISCUSSION

We estimate that the sea scallop, Placopecten magellanicus, resource in the off-shore waters of the Northeastern United States contain approximately 16 billion individuals, representing approximately 217,520 mt tons of scallop meats. The sea scallop biomass on Georges Bank in United States waters alone is 3.5 times greater than the average biomass from 1977 to 1988 for all of Georges Bank (31,500 mt tons, Canadian and United States portions combined; McGarvey et al, 1992). In the previous fishing year (March 2002 to February 2003) the sea scallop fishery harvested 23,237 mt tons of sea scallop meats, equivalent to 11% of the biomass surveyed in the 2003 survey (NMFS Fishery Statistics Report).

The high biomass is primarily a result of high densities and very large scallops within the closed areas of the mid-Atlantic and Georges Bank. However, sea scallops abundances have increased in the open areas as well.

The increase in numbers of scallops in the mid-Atlantic is the result of an extremely large recruitment of juvenile scallops. Periodically large year classes of sea scallops have occurred throughout the history of the fishery and are believed to result from favorable environmental conditions (Serchuk et at. 1979, Caddy 1989), Caddy (1989) states that scallop biomasses have shown wide fluctuations from year to year that are not all the result of fishing. The sea scallop fishery falls outside the stable classification and represents a cyclical, irregular or spasmodic fisheries time series depending on the geographic location of the bed (Caddy & Gunand 1983, Caddy 1989).

The high abundance of large scallops within the closed areas may be producing many more larval scallops than in previous years. Sinclair et al. (1985), Caddy (1989) and McGarvey et al. (1992) emphasize the importance of larger size on the fecundity-per-recruit relationship. McGarvey et al. (1993) suggest that a stock-recruitment relationship exists for +5 y old sea scallops but that the fishery removes the majority of scallops greater than age 3.25 y. The sea scallop densities and sizes we observed exceed previously recorded densities and sizes over large geographic areas (Stokesbury 2002).

The high abundance of large scallops within the closed areas on Georges Bank may also be inhibiting recruitment. Few prerecruit scallops (<90-mm shell height) were observed in these closed areas. Replacement of the adult scallop stock is barely occurring in the Nantucket Lightship Area and Closed Area II with the prerecruit densities observed in 2003, using an instantaneous mortality rate of 0.10 and assuming a 17 y life span (Caddy 1989). If the instantaneous mortality rate is slightly higher than 0.10, for example 0.12, then the 2003 recruits would not sustain the adult scallop population at its present density.

A sea scallop mass-mortality event may be occurring in the southern portion of Closed Area II due to sea star predation. Here the highest densities of large sea stars and dead scallops occurred. The NMFS observed a high instantaneous mortality rate (M = 0.6) in this area although there is considerable uncertainty in the estimate due to tire year-to-year variability of the survey (the mean number of scallops per tow was 1255 in 2002 and 698 in 2003, >40 and >70 mm shell heights, respectively; D. Hart, personal communication). Between 1928 and 1961 nine mass moralities occurred in the Gulf of St. Lawrence populations (Dickie & Medcof 1963). The more devastating widespread mass moralities (up to 80% of the population) were attributed to water temperatures exceeding the upper lethal limits for scallops. The less severe mass moralities (up to 25%) were attributed to sea star predation (Dickie & Medcof 1963).

Sea stars were twice as abundant as sea scallops in 2003, although many were very small suggesting a new recruitment. In our previous surveys of Georges Bank and the Mid-Atlantic sea scallops far outnumbered sea stars. Therefore, localized high natural mortality caused by sea star predation may be an increasing concern for the sea scallop fishery. We are examining predator-prey distributions and interactions and spatial variation in natural mortality.

The video survey technique is fast, accurate, precise, and provides information on the biology of scallops and the associated habitat without disturbing the sea floor (Stokesbury 2002). The two-stage sampling technique and the unaligned systematic sample provided similar means and low coefficients of variation compared with many invertebrate surveys (Krebs 1989).

The 2003 video survey was a success due to the strong collaboration and participation of the sea scallop fishing fleet and supporting industries. Preliminary data from this survey have been presented Io the New England Fisheries Management Council and the NMFS. A closure that will protect the area of high sea scallop recruitment in the southern portion of the Hudson Canyon closed area extending south into open waters has been proposed in the recent sea scallop fishery management plan. Further, the southern portion of Closed Area II may be opened earlier than scheduled due to the high natural mortality observed in this area.

TABLE 1.
The sea star, Asteroidea, density (sea star * [m.sup.-2]), number of
stations sampled, standard error (SE), coefficient of variation (CV),
and area surveyed ([km.sup.2]) using a two-stage sampling design.

                                Sea Stars

Areas          Stations   Mean [m.sup.2]    SE

Georges Bank
  NLSA            128          1.67        0.254
  CAI              97          0.09        0.023
  CAII            186          0.13        0.050
  Open            513          0.34        0.032
  Sum             924
Mid-Atlantic
  HC              160          0.97        0.131
  VA               34          0.46        0.111
  Open            657          1.02        0.063
  Sum             851
  Total          1775

                Sea
               Stars

                                    Millions of
Areas           CV%    [km.sup.2]    Sea Stars

Georges Bank
  NLSA         15.19      3951           262
  CAI          26.76      2994           740
  CAII         38.88      5742          5387
  Open          9.30     15836          6618
  Sum                    28523         13007
Mid-Atlantic
  HC           13.47      4939          4793
  VA           23.99      1050           487
  Open          6.21     20281         20673
  Sum                    26270         25952
  Total                  54793         38959

Estimates were calculated for the Nantucket Lightship Area (NLSA),
Closed Area I (CAI), Closed Area II (CAII), Hudson Canyon (HC), and
Virginia Beach (VB) closed areas and open areas of Georges Bank and
the mid-Atlantic.

TABLE 2.
The sea scallop, Placopecten magellanicus, density (scallop *
[m.sup.-2]), number of stations sampled, standard error (SE),
coefficient of variation (CV), using a two-stage sampling
design and an unaligned systematic sample.

                                 Two-Stage Design

Areas          Stations   Mean [m.sup.2]    SE     CV%

Georges Bank
  NLSA            128          0.21        0.040   19.4
  CAI              97          0.16        0.036   22.1
  CAII            186          0.26        0.045   17.4
  Open            513          0.09        0.008    9.3
  Sum             924
Mid-Atlantic
  HC              160          0.79        0.250   31.5
  VA               34          0.04        0.008   21.1
  Open            657          0.40        0.076   19.2
  Sum             851
  Total          1775

                    Unaligned Design

                                                  Meat
Areas          Mean [m.sup.2]    SE     CV%    Weight (g)

Georges Bank
  NLSA              0.21        0.029   14.0      39.8
  CAI               0.16        0.021   12.7      30.7
  CAII              0.26        0.021    8.2      33.2
  Open              0.09        0.007    7.4      14.5
  Sum
Mid-Atlantic
  HC                0.79        0.064    8.0       9.1
  VA                0.04        0.017   42.9      13.1
  Open              0.40        0.024    6.1       7.9
  Sum
  Total

                            Millions of   Scallop Meats
Areas          [km.sup.2]    Scallops        mt tons

Georges Bank
  NLSA            3951           816          32465
  CAI             2994           489          15016
  CAII            5742          1474          48980
  Open           15836          1436          20823
  Sum            28523          4215         117285
Mid-Atlantic
  HC              4939          3915          35714
  VA              1050            41            533
  Open           20281          8075          63988
  Sum            26270         12030         100235
  Total          54793         16245         217520

Meat weight (g) estimates, based on the shell height frequency
observed in each area and the shell height/meat weight regression,
and area surveyed ([km.sup.2]), provided the metric tons of meat
weight in each area. Estimates were calculated for the Nantucket
Lightship Area (NLSA), Closed Area I (CAI), Closed Area II (CAII),
Hudson Canyon (HC) and Virginia Beach (VB) closed areas and open
areas of Georges Bank and the mid-Atlantic.

ACKNOWLEDGMENTS

The authors thank B.J. Rothschild for his support and guidance. The authors thank the owners, captains, and crews who sailed with us on the F/V Friendship, F/V Huntress, E/V Liberty, and F/V Nordic Pride. M. Steinell (Deep Sea Light & Power), D. Bentley (Cortland Cable), D. Chambers (Maritime Hydraulics), and P. Clingman (Diversified Marine) helped us design and build the tension sensitive hydraulic winch, cables and sampling pyramid. P. Christopher (NMFS) and L. Gavlin (USCG) provided the Letters of Authorization and insured smooth transitions between fishing and surveying for the vessels. P. Rago (NMFS) and T. Truong provided helpful comments on sampling variance. M. Stokesbury and B. Rothschild reviewed and provided many helpful comments, which greatly improved the manuscript. Aid was provided by SMAST, the Massachusetts Department of Marine Fisheries, and the sea scallop fishery and supporting industries.

LITERATURE CITED

Brand, A. R. 1991. Scallop ecology: distributions and behaviour. In S. E. Shumway, editor. Scallops: biology, ecology and aquaculture. Amsterdam: Elsevier, pp. 517-584.

Caddy, J. F. 1989. A perspective on the population dynamics and assessment of scallop fisheries, with special reference to sea scallop, Placopecten magellanicus (Gmelin). In: J. F. Caddy, editor. Marine invertebrate fisheries: their assessment and management. New York: John Wiley & Sons, pp. 559-589.

Caddy, J. F. & J. A. Gulland. 1983. Historical patterns of fish stocks. Mar. Pol. 7:267-278.

Cochran, W. G. 1977. Sampling techniques. 3rd ed. New York: John Wiley & Sons, pp. 330.

Dickie, L. M. & J. C. Medcof. 1963. Cause of mass mortalities of scallops (Placopecten magellanicus) in the southwestern Gulf of St. Lawrence. J. Fish. Res. Bd. Canada 20:451-482.

Glantz, S. A. 2002. Primer of biostatistics. New York: McGraw-Hill, pp. 489.

Hart, D. R. 2001. Individual-based yield per recruit analysis, with an application to the Atlantic sea scallop, Placopecten magellanicus. Can. J. Fish. Aquat. Sci. 58:2351-2358.

Hart, D. R. 2003. Yield- and biomass-per-recruit analysis for rotational fisheries, with an application to the Atlantic sea scallop Placopecten magellanicus. Fish. Bull. 101:44-57.

Krebs, C. J. 1989. Ecological methodology. New York: Harper & Row, pp. 654.

McGarvey, R., F. M. Serchuk & I. A. McLaren. 1992. Statistics of reproduction and early life history survivial of the Georges bank sea scallop (Placopecten magellanicus) population. J. Northw. Atl. Fish. Sci. 13: 83-99.

McGarvey, R., F. M. Serchuk & I. A. McLaren. 1993. Spatial and parentage analysis of stock-recruitment in the Georges bank sea scallop (Placopecten magellanicus) population. Can. J. Fish. Aquat. Sci. 50:564-574.

Murawski, S. A., R. Brown, H.-L. Lai, P. J. Rago & L. Hendrickson. 2000. Large-scale closed areas as a fishery-management tool in temperate marine ecosystems: the Georges Bank experience. Bull. Mar. Sci. 66: 775-798.

Rago, P., S. Murawski, K. Stokesbury, W. DuPaul & M. McSherry. 2000. Integrated management of the sea scallop fishery in the northeast USA: research and commercial vessel surveys, observers, and vessel monitoring systems ICES Mar. Sci. Symp. C[M.SUP.2]000/W:13. pp. 18.

Serchuk, F. M., P. W. Wood, J. A. Posgay & B. E. Brown. 1979. Assessment and status of sea scallop (Placopecten magellanicus) populations off the Northeast coast of the United States. Proc. Nat. Shellfish Assoc. 69:161-191.

Sinclair, M., R. K. Mohn, G. Robert & D. L. Roddick. 1985. Considerations for the effective management of atlantic scallops. Can. Tech. Rep. Fish. Aquat. Sci. 1382:1-99.

Stokesbury, K. D. E. 2000. Physical and biological variables influencing the spatial distribution of the giant scallop, Placopecten magellanicus. In: A workshop examining potential fishing effects on population dynamics and benthic community structure of scallops with emphasis on the weathervane scallop Patinopecten caurinus in Alaskan waters. Alaska Department of Fish and Game, Division of Commercial Fisheries, Juneau. Special Publication 14, pp. 13-19.

Stokesbury, K. D. E. 2002. Estimation of sea scallop abundance in closed areas of Georges Bank. USA. Trans. Am. Fish. Soc. 131:1081-1092.

Stokesbury, K. D. E. & J. H. Himmelman. 1993. Spatial distribution of the giant scallop Placopecten magellanicus in unharvested beds in the Baie des Chaleurs, Quebec. Mar. Ecol. Prog. Ser. 96:159-168.

Thouzeau, G., G. Robert & S. J. Smith. 1991. Spatial variability in distribution and growth of juvenile and adult sea scallops Placopecten magellanicus (Gmelin) on eastern Georges Bank (Northwest Atlantic). Mar. Ecol. Prog. Ser. 74:205-218.

KEVIN D. E. STOKESBURY, * BRADLEY P. HARRIS, MICHAEL C. MARINO II AND JACOB I. NOGUEIRA

School for Marine Science and Technology, University of Massachusetts Dartmouth, 706 South Rodney French Boulevard, New Bedford, Massachusetts, 02744-1221

* Corresponding author. Phone: [phone]; Fax: [phone]; E-mail: [e-mail]