A note on a spawner--recruit relationship for a heavily exploited bivalve: the case of northern quahogs (hard clams), Mercenaria mercenaria in great south bay New York.

ABSTRACT The Town of Islip, NY has collected a long-term data set (1977-2004) on northern quahog (hard clam), Mercenaria mercenaria, abundance. The data comprise approximately 350 duplicate 1 [m.sup.2] samples each year taken with a clam shell bucket. All samples were sieved through a 6.4 mm sieve and the hard clams enumerated by size. In addition, clam landings data for the town waters are available for the same time period. Clam populations have declined from their peaks in the 1970s to very low levels in the 1990s and 00's. These dramatic shifts in population abundance have made the exploration of spawner/recruit relationships possible. A number of alternate models were attempted, but based on knowledge of the biology of the species and other factors, all but two did not appear to be plausible. The two models (2nd order polynomial and Log) yielded high [r.sup.2] values and intercepted the 0 axis between 0.5 and 0.75 adult clams [m.sup.-2] indicating a density dependent effect on recruitment. The polynomial model also suggested a carrying capacity level of about 5 adult clams [m.sup.-2] and a density dependent upper level of density. This is the first time a spawner recruit relationship has become apparent for hard clams.

KEY WORDS: northern quahog, hard clam, Mercenaria mercenaria, recruitment, pl recruit

INTRODUCTION

Because of their high fecundity, exploited populations of bivalve molluscs are generally believed not to be subject to traditional spawner/recruit (SR) constraints (see Hancock 1973 for a general review: cockles, Cardium edule [Hancock & Urquart 1965]; surfclams, Spisula solidissima; ocean quahogs, Arctica islandica [Murawski & Serchuk 1989] the American oyster, Crassostrea virginica [MacKenzie 1996]) and thus are "particularly resistant to overfishing" (Jamieson 1993). A notable exception is the sea scallop, Placopecten magellanicus, populations on Georges Bank where McGarvey et al. (1993) showed a significant correlation between egg production and recruitment. The presence of a spawner/recruit relationship (SRR) has also been debated for some highly fecund fish species (see a review by Myers & Barrowman [1996] and a discussion by Haddon [2001]). In spite of few demonstrated relationships it is obvious that an SRR must exist at some low stock level, and although SRRs are considered to be of critical importance in management of most exploited fish, and many crustacean populations (Myers & Barrowman 1996, Wahle 2003, Zheng & Kruse 2003) they have rarely been considered for bivalves. The fundamental problem for bivalves is that the relationship is easily obscured over large ranges of population density by, suboptimal conditions during larva life (Powell et al. 2004, Hofmann et al., 2004, Olafsson et al. 1994), large populations, environmental factors influencing fecundity, and at least for Mercenaria mercenaria, the hard clam, postsettlement losses to predation (Malinowski 1993, Kraeuter 2001). It is often only when sufficient data have been collected that include at least low and moderate stock levels that such a relationship can be elucidated.

Classic fishery management models of Ricker (1954) and Beverton and Holt (1957) typically assumed the relationship between stock and subsequent recruitment was negatively density-dependent. Recent publications have indicated that, despite high fecundity, invertebrates that are subject to high levels of exploitation such as tridacnid clams (Munro 1989) and some scallops (Caddy 1989) may be limited by density-dependent factors that are positive. Munro (1989) suggested that under high exploitation some tridacnid clams may become so dispersed that spatial patterns may limit spawning success because of dilution of sperm.

Such a mechanism was proposed for hard clams in the early part of the 20th century (Belding 1912). Caddy (1989) suggested a similar cause for limits of some scallop stocks. Experimental evidence for such relationships in bivalves does not exist, but Grant et al. (1998) estimated natural fertilization rates during mass spawnings in the field for Paphies austral&. When animals were aggregated (females within 1 m of males) fertilization was 100%, but in areas where females were isolated (>1 m from males) fertilization rates dropped to 83%. Levitan et al. (1992) clearly demonstrated that group size, degree of aggregation and flow velocity could affect fertilization success in the sea urchin Strongylocentrotus franciscanus. This study also demonstrated that the effects took place on a scale of a meter to a few meters. Further research on other invertebrates (Thomas 1994, Babcock et al. 1994) has confirmed this general conclusion, although the particulars may be strongly species dependant. Yund and Meidel (2003) have recently shown that other factors such as release of gametes in viscous fluids may mitigate many of the processes formerly hypothesized, and serve to make echinoderms much less sperm-limited than previously demonstrated. Yund (2000) has suggested that in natural populations data indicate that mechanisms have developed that reduce the potential effects of sperm limitation.

Peterson (2002) examined the numbers of recruiting hard clams in data sets from various experiments that took place from the late 1970s to 2001 in North Carolina. This effort reviewed the general information from fishery increases and declines, and resampled previously sampled experimental areas to makes a strong case for recruitment overfishing on hard stocks in North Carolina. Unfortunately, there are rarely sufficient annual time series data on hard clams to evaluate the effects of stock declines in other areas or to determine if a SRR might exist.

STUDY AREA

The hard clam (Mercenaria mercenaria) supported a large fishery in Great South Bay, New York (GSB), that at one time produced nearly 50% of the entire east coast landings. Because of its importance in the local economy, substantial data have been collected. The fishery landings data clearly indicate a decline in abundance (Fig. 1). A fishery independent survey of abundance has been conducted by the Town of Islip and these data follow a downward trend similar to that of the fishery data (Fig. 2), and landings data from the Town of Islip clearly mirror the decline of the fishery (Buckner unpublished data) (Fig. 1).

[FIGURES 1-2 OMITTED]

A number of hypotheses have been put forth to explain the decline in abundance, but it is clear that this stock was being exploited beyond its capacity to recruit in the late 1970s and early 1980s (Fig. 3).

[FIGURES 3 OMITTED]

We have been developing a numerical simulation model to investigate the effects of various factors on the biology of the hard clam. In this effort, we have been fortunate to have the extensive data set collected by the Town of Islip that provides direct fishery-independent measurements of spawners and recruits for an extensive area in GSB. In preparation for the model, analysis of the data for a potential SRR was required.

There are a number of ways that the term recruit has been used, for the purposes of this study we examine the correlations between the average numbers of adults [m.sup.-2] and the average numbers of 1-year-old clams [m.sup.-2] (recruits) with the appropriate lag time. Later we present the results of that analysis, and the first demonstration of an SRR for a hard clam population. It is important to note that we are utilizing average numbers over a large area and are focusing on a clam population that has been at a density <5 [m.sup.-2] since the late 1970s.

DATA

The data from the town of Islip has been used as representative of the general conditions in GSB, and data on landings (Fig. 1) indicate that it is representative of the bay as a whole. Islip has about 6,000 hectares of bay bottom. Of this 80% is open for harvest of clams, 10% is closed and 13% is leased. Since 1977 Islip has been conducting an annual fishery-independent survey of its clam populations (Buckner 1984). Sampling methodology has remained relatively unchanged since the survey's inception. All Islip waters are divided into 400 [m.sup.-2] grids. Approximately 350 grids were sampled each year (Buckner 1984). Duplicate samples are removed from a randomly chosen section within each grid. The survey is conducted using a barge with a clam shell bucket that removes about 1.02 [m.sup.-2] of bottom. Materials from all duplicate samples at each station are separately sieved through a 6.4 mm sieve, all live and dead clams are enumerated, and all live clams are measured. Clams of age 0, 1 and 2, can generally be separated from other year classes based on size-frequency distributions. From the thickness measurements, clams were separated into seed and various commercial size classes. The latter are 25.4-36.4 mm (littleneck), 36.5-41.3 mm (cherrystone) and >41.3 mm (chowder). From time to time specimens have been set aside for shell sectioning and aging of the animals, but for purposes of this study clams 2 y old and older have been lumped into one category representing the spawning stock. These data clearly show a decline in hard clam stocks beginning at the time of the initiation of the survey (Fig. 2).

DATA ANALYSIS

Data from the 0 y class were eliminated from the analysis because the use of a 6.4 mm sieve may allow an unknown fraction of individuals smaller than 6.4 mm to escape. To avoid any possible bias caused by missing animals, we have concentrated our efforts on recruitment of age 1 individuals. The data from each year of the survey have been reduced to represent mean numbers [m.sup.-2] for age 1 (9-17 mm in thickness) and those >age 1. The annual average number of recruits and adults in Islip waters were entered in an Excel spread sheet, plotted and then subjected to Trendline analysis to evaluate which of the functions (linear, natural log, 2nd order polynomial, power) gave the best fit and the most realistic shape to the curve (Figs. 4, 5 and 6). Early evaluations were conducted on data from 1977-2000, and were used in model development. Subsequent to these analyses we received additional data, and we decided to evaluate the new data sequentially as it appeared. In addition, we attempted to fit a Ricker and Beverton Holt type curve to data in such a way that the residual error was at the approximate lowest level (Haddon 2001).

[FIGURES 4-6 OMITTED]

RESULTS

Since 1999 the average density of clams has dropped from 1.82-1.04 clams [m.sup.-2] and adult clams have decreased from 1.250.86 clams |[m.sup.-2]. Neither the Ricker curve nor the Beverton Holt curve provided a better fit to the data than simple linear, log, polynomial or power functions. The Ricker curve and all four of the latter models generally show an increasing [r.sup.2] with increasing data representing low population levels (Table 1).

Linear models yield the lowest correlation coefficients in each year (Table 1) and in the 1999 to 2001 period. This model yields the highest adult population density at which 0 recruits are predicted, and shows the greatest change in this parameter with time (Table 2). These curves also predict what seems to be an unrealistic rate of increasing recruitment with increasing population size (Fig. 4).

Table 2. Approximate adult density [m.sup.-2] at which various spawner recruit curves indicate insufficient adults to provide for a sustainable population.

The point at which recruitment is 0 for Log models remains stable for the 2000 to 2003 period (Table 2), and whereas the recruitment rate continues to increase with adult density, the rate of increase begins to asymptote (Fig. 4 and 5) at a density of 30 clams [m.sup.-2].

The low adult density of 0 recruitment for the polynomial models is about the same as for the log models (Table 2), but these models predict a maximal point of recruitment at about 5 clams [m.sup.-2] and 0 recruitment by adult population density of about 9 clams [m.sup.-2] (Figs. 4 and 6, Table 3).

The power function and the linear model (Fig. 4) suffer from the same shortcomings in that they imply rapidly increasing recruitment with stock size, a relationship that at some point must become untenable. In the current case, the Ricker curve also implies rapidly increasing recruitment with stock size and was eliminated from consideration. Both the natural log and 2nd-order polynomial approximations provide reasonable fits to the data and have high [R.sup.2] (Figs. 4, 5 and 6). All models imply that a density of spawning-stock exists at which the density of clams is insufficient to maintain the population (Table 2). Projections of these curves beyond the upper limits of the data (approximately 5 adults [m.sup.-2]) (Figs. 5 and 6) indicate that either the natural log or polynomial model could provide recruitment levels that mimic most data from hard clam populations (See review by Fegley [2001] for density of natural hard clam populations). Both of these models yield high [r.sup.2] values indicating they are statistically robust estimates, but the data do not allow us to distinguish which of these best describes the natural condition for hard clams.

DISCUSSION

It has been shown that simple modeling of the hard clam population data from Great South Bay, New York suggests an SSR exists; there are a number of possible curves that could explain the observed data, and most have a relatively high correlation coefficient (Table 1), but we are unable to determine whether a natural log or second-order polynomial provides the best fit of the data. Fegley (2001) reviewed the population data on hard clams and found reported densities as high as 856 [m.sup.-2], but most populations had standing stocks of <50 [m.sup.-2]. Larger population densities probably represent collections in smaller areas or in areas of known commercial populations, and thus do not represent a bay or waters of a town. Hard clam population density, based on field sampling, may be >500 [m.sup.-2] over areas of a few [m.sup.-2], but typical density is log normal in distribution and in the range of 4-8 [m.sup.-2] with 80% of published accounts lying in the range of 1-15 [m.sup.-2] (Fegley 2001). Surveys of large areas in the southern New England through the MidAtlantic area (Table 4) have reported densities of adult hard clams that are mostly within the range of the polynomial model derived from the Islip data. All these populations have been subject to harvest, and with the exception of the Metedeconk River, and the Maryland bays after the 1950s could be derived from either the log or polynomial models. Salinity in the Metedeconk River is generally below the optimum for hard clams and may provide an explanation for the discrepancy with the model. The data from the Maryland bays was collected by slightly different methods (hydraulic escalator dredge) in recent years, but there is no evidence that this gear is inappropriate for estimating density of hard clams. It is possible that the model, derived in colder waters of GSB overestimates the average population density required to maintain a population in more southern waters. This would suggest that southern clams have either a greater reproductive output or recruitment success than more northern populations. How the lower average population density required for population maintenance would affect the shape of the curve and the estimates of maximal average population densities cannot be evaluated from present data. It is also likely that different models may be required for populations in bays with differing temperature, salinity, flushing rates and predator suites.

Recruitment from the larval stage has rarely been measured for hard clams and no data are available for animals <1 mm (Fegley 2001). A few studies have followed 0 age animals and localized settlement of hundreds [m.sup.-2] have been reported, but even in these "best cases" seed numbers have been reduced to <100 [m.sup.-2] by the end of the first winter (Connell 1983). There are more studies that have sampled benthos in areas of commercial hard clam populations, and hard clams in the range of 5-20 mm shell length would certainly be adequately sampled by typical grab or box core methods. Only a few reports exist in the benthic or other literature where juvenile hard clams are found in abundances of >1-5 [m.sup.-2]. Experimental plantings of small hard clams suffer significant loss unless they are protected from predators (Kraeuter & Castagna 1989). We have used these general parameters to evaluate the predictions of the various classes of curves derived from the Islip data. Log models suggest that recruitment begins to quickly asymptote above population levels of 30 [m.sup.-2], and the 2nd-order polynomial model suggests that negative density-dependent factors become important above approximately 5 adult individuals [m.sup.-2] (Table 3). One's initial reaction is that the 2nd-order polynomial expression does not fit existing data for hard clam populations because it implies that very dense populations of hard clams cannot exist. The few surveys that examine bay-wide data (all of these are in areas of fished populations) conform reasonably well to the model predictions (Table 4), however the extent of the very densest aggregations cited in the review by Fegley (2001) may be much more limited. Thus the spatial scale at which the data are collected may have a strong influence on the model output and its interpretation.

Use of the x-axis intercept method assumes that the population is dispersed equally across the bottom, a situation that never occurs for wild populations. Alternatively, spawner/recruit curves could be forced through the origin at a density of 0. This may be a reasonable assumption for some motile species, but it seems theoretically plausible that organisms such as clams that are relatively fixed in place and rely on broadcast spawning for fertilization, could become so dispersed that reproduction would cease. One might expect that the SR curve would cross the y-axis at approximately the density required to assure fertilization. Our models suggest that under current conditions for GSB this density is about 0.75 clam [m.sup.-2]. The most convincing evidence for such an effect was provided by Levitan et al. (1992) for sea urchins. This experimental study simulated spawning and measured fertilization success at distances ranging from 0.5-2 m, with different degrees of aggregation of the adults (0.25-4 individuals [m.sup.-2]) and at different current speeds. Their results clearly indicated that density of adults, distances between adults and current velocity can have substantial negative density-dependent effects on fertilization success. These effects become significant at densities of 1 [m.sup.-2], approximately the same scale our models suggest that hard clams may begin to have difficulty sustaining their populations. The limited data presented by Grant et al. (1998) for the bivalve Paphies suggest the possibility of such a limitation for clams. Some species, such as the starfish Acanthaster planci are able to reproduce at lower densities (Babcock et al. 1994). Yund and Meidel (2003) have shown that those echinoderm species that have gametes in viscous fluids may have developed mechanisms to extend the life of eggs and sperm, prevent their dilution and compensate for the effects of releasing them in a medium that tends to reduce fertilization success. Experimental data relating fertilization spatial patterns and densities are not yet available for hard clams. It is known from hatchery work that the presence of sperm can trigger release of eggs in many bivalve species. This implies that mechanisms to increase fertilization success are used to some extent by some bivalves.

Whether or not the curves generated are sufficiently realistic to use in a predictive fashion, it is important to realize that under low stock levels some form of SRR exists. It is also important to remember that these curves are generated from natural populations that are being fished, and thus may be compromised to some degree for the potential density effects. Fishing on stocks approaching the average observed low-density level may foster conditions that exacerbate the density effects, because fishing concentrates on the densest portions of the population. Thus, fishing may serve to increase the distances between individuals even before it appreciably lowers overall population levels.

At the other end of the spectrum, the log curves indicate a steadily increasing recruitment with increasing stock, but the 2nd-order polynomial indicates a classic negative density dependency similar to that typically seen for managed fish and crustacean populations (Ricker 1975, Garcia 1988). Whether negative density-dependent factors equivalent to the compensatory effects hypothesized for fish populations exist in natural bivalve populations is unknown, but growth inhibition and implied reductions in fecundity, have been found in high densities of some bivalve species (Powell et al. 1987, Powell et al. 1995). Negative effects caused by the filtration activity of adults destroying their own and other species larval stages have been shown for a number of species including hard clams (Kristensen 1957, Andre & Rosenberg 1991, Kurkowski 1981). In addition, Rice et al. (1989) and Kennish (1978) suggested that reducing adult density increased recruitment at high clam density. Malinowski (1993) described experiments in which adult hard clams were experimentally placed at densities of 4-67 per 0.33 [m.sup.2] and examined meat displacement as a surrogate for condition. Only clams held at the highest density showed a statistically significant reduction in meat weight. Despite these observations, hard clam populations are typically at densities too low for significant negative density-dependent effects to occur (Best 1978, Young & Young 1978, Maurer 1983, Malinowski 1993). In densely planted aquaculture plots with high survival Kraeuter and Castagna (1985) reported growth inhibition during the second year of field life.

In summary, it has been shown that simple stock recruitment curves can be derived at least when low (~1 clam [m.sup.-2]) adult population densities of hard clams such as those currently found in GSB, New York are present. These curves suggest that in GSB under current conditions recruitment is limited at clam population densities below about 0.75 individual [m.sup.-2]. The most likely explanations seem to be that fecundity is reduced or spawning and fertilization success or post fertilization survival are hindered under current environmental conditions. This effect could become a factor limiting, or at least greatly reducing, recovery of the populations. Although not unequivocal, some curves suggest negative density-dependent effects at high population densities. Plausible mechanisms exist, but as with the positive density-dependent effects, little empirical data exist that can confirm or refute these hypotheses for most bivalves and for hard clams in particular.

Data from most hard clam populations suggest low recruitment to age 1 when compared with other commercial bivalves. When the current data are placed in the context of a recruit/adult basis (Fig. 7) data from Islip show a drop in recruitment success beginning in 1998 when compared with prior decades. This recent drop in recruitment cannot be ascribed to over fishing, and may be influenced by brown tide (Aureococcus anophagefferens) that has become a regular feature of this estuary. Whatever the cause or causes the data indicate that replacement time for adults in the 1996 to 2003 time period is 2.6 times longer than it was from 1979-1995.

[FIGURE 7 OMITTED]

ACKNOWLEDGMENTS

The authors thank the individuals who comprised the field crews who collected the data over the years and to Marie Zelinski who compiled much of the information. We also thank the Town of Islip, New York and Suffolk County for funding the collection of data on clams and water quality throughout the years. This study was funded, in part, by New York Sea Grant.

LITERATURE CITED

Andre, C. & R. Rosenberg. 1991. Adult-larva interactions in the suspension feeding bivalves Cerastoderma edule and Mya arenaria. Mar. Ecol. Prog. Ser. 71:227-234.

Babcock, R. C., C. N. Mundy & D. Whitehead. 1994. Sperm diffusion models and in situ confirmation of long-distance fertilization in the free-swimming asteroid Acanthaster planci. Biol. Bull. 186:17-28.

Belding, D. L. 1912. The quahog fishery of Massachusetts. Comm. Massachusetts Dept. Conserv., Div. Fish and Game. Mar. Fish Ser. No. 2. 41 pp.

Best, B. A. 1978. The effects of suspension feeding by the bivalve, Mercenaria mercenaria, on community structure. MS. Thesis, University of Florida. Gainesville, Florida. 40 pp.

Beverton, R. J. H. & S. J. Holt. 1957. On the dynamics of exploited fish populations. Fish. Investig. Ser. II. Mar. Fish. G.B. Minist. Agric. Fish. Food 19:1-533.

Buckner, S. C. 1984. Aspects of the population dynamics of the hard clam, Mercenaria mercenaria L., in Great South Bay, New York. Ph.D. Diss. SUNY Stony Brook, New York. 217 pp.

Caddy, J. F. 1989. A perspective on population dynamics and assessment of scallop fisheries, with special reference to the 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.

Cole, R. W. & L. E. Spence. 1977. Hard clam survey of Rehoboth and Indian River Bays. Annual Report to NMFS. Contract No. 04-4043-466. pp. 1-30.

Conwell, Jr. R. 1983. Seasonal mortality and population dynamics of juvenile hard clams, Mercenaria mercenaria L., on an entertidal mudflat in New Jersey. MS Thesis Rutgers Univ.

Fegley, S. R. 2001. Demography and dynamics of hard clam populations. In: J. N. Kraeuter & M. Castagna, editors. Biology of the hard clam. New York: Elsevier Science.

Garcia, S. 1988. Tropical penaeid prawns. In: J. A. Gulland, editor. Fish population dynamics. 2nd edition. New York: John Wiley and Sons. pp. 219-244.

Grant, C. M., S. H. Hooker, R. C. Babcock & R. G. Creese. 1998. Synchronous spawning and reproductive incomparability of two bivalve species: Paphies subtriangulata and Paphies australis. Veliger 41 : 148-156.

Haddon, M. 2001. Modeling and quantitative methods in fisheries. New York: Chapman and Hall. 406 pp.

Hancock, D. A. 1973. The relationship between stock and recruitment in exploited invertebrates. Rapp. P.-V. Reun. Cons. Perm. Int. Explor. Mer. 164:113-131.

Hancock, D.A. and A.E. Urquart. 1965. The determination of natural mortality and its causes in an exploited population of cockles (Cardium edule L.) Fish. Invest. (London) 2 24:1-40.

Hofmann, E. E., E. N. Powell, E. A. Bochenek & J. M. Klinck. 2004. Critical conditions for larval success: influence of environment and food supply on survival of Crassostrea gigas larvae: A modeling study. ICES J. Mar. Systems 61(4):596-616.

Humphries, E. M. & F. C. Daiber. 1968. Shellfish survey of Indian River and Rehoboth Bay, Delaware. Tech. Rept. to US Public Health Service, Northeast Marine Health Science Laboratory. Delaware Project PH-86-66-159.70 pp.

Jamieson, G. S. 1993. Marine invertebrate conservation: evaluation of fisheries over-exploitation concerns. Am. Zool. 33:551-567.

Joseph, J. W. 1986. Inventory of New Jersey's estuarine shellfish resources. July 1, 1985 to June 30, 1986. PL 88-309, US DOC, NOAA, NMFS. December 12, 1986. 35 pp. plus attachments.

Joseph, J. W. 1987. Inventory of New Jersey's estuarine shellfish resources. July 1, 1986 to August 30, 1987, PL 88-309, US DOC, NOAA, NMFS. December 9, 1987.79 pp.

Joseph, J. W. 1989. Inventory of New Jersey's estuarine shellfish resources. Completion Report for July 1, 1985 to June 30, 1989, PL 88-309, US DOC, NOAA, NMFS. September 1989. 75 pp.

Kennish, M. J. 1978. Effects of thermal discharges on mortality of Mercenaria mercenaria in Barnegat Bay, New Jersey. Environ. Geol. 2: 223-254.

Kraeuter, J. N. and M. Costojna. 1985. The effects of seed size, shell bags, crap drops and metting on the survival of the northern hard clam Mercenaria mercenaria (L.). J Shellfish Res. 5:69-72.

Kraeuter, J. N. and M. Costojna. 1989. Factors affecting the growth and survival of clam seeds planted in the natural environment. Ch 6. In: Manzi, J. J. and M. Costojna (eds.) Clam manufacture in North America. Developments in Aquaculture and Fisheries Science 19:149-165.

Kraeuter, J. N. 2001. Predators and predation In: J. N. Kraeuter & M. Castagna, editors. Biology of the hard clam. New York, New York: Elsevier. pp. 441-568

Kristensen, I. 1957. Differences in density and growth in a cockle population in the Dutch Wadden Sea. Arch. Neel. Zool. 12:351-453.

Kurkowski, K. P. 1981. Effects of filtration by adult Mercenaria mercenaria upon its larvae. MS. Thesis, State University of New York, Stony Brook, New York. 70 pp.

Lazar, N., A. Ganz & A. Valliere. 1999. Quahaug stock assessment and implementation of an interim management plan in Greenwich Bay, Rhode Island. In: A. Ganz, A. Valliere, M. Gibson & N. Lazar, editors. Narragansett Bay quahaug management plan. RI Department of Environmental

Management, Div. Fish and Wildlife, Coastal Fisheries Laboratory, Wakefield, RI.

Levitan, D. R., M. A. Sewell & F.-S. Chia. 1992. How distribution and abundance influence fertilization success in the sea urchin Strongylocentrotus franciscanus. Ecology 73:248-254.

MacKenzie, C. L. 1996. Management of natural populations. In: V. S. Kennedy, R. I. E. Newell, & A. F. Eble, editors. The eastern oyster, Crassostrea virginica. University of Maryland Sea Grant, College Park, MD.

Malinowski, S. M. 1993. Effects of density on adult growth, survivorship, fecundity and recruitment of quahogs. In: M. A. Rice & D. Grossman-Garber, editors. Proceedings of the second Rhode Island shellfish industry conference. Rhode Island Sea Grant. pp. 79-90.

Maurer, D. 1983. The effect of an infaunal suspension feeding bivalve Mercenaria mercenaria (L.) on benthic recruitment. P.S. Z.N. Mar. Ecol. 4:263-274.

McCloy, T. W. & J. W. Joseph. 1983. Inventory of New Jersey's estuarine shellfish resources. July 19, 1982 to June 30, 1983. PL 88-309, US DOC, NOAA, NMFS. November 28, 1985. 25 pp. plus attachments.

McCloy, T. W. & J. W. Joseph. 1985. Inventory of New Jersey's estuarine shellfish resources. Completion Report for the period December 1, 1979 to June 30, 1985. PL 88-309, US DOC, NOAA, NMFS. December 1985.28 pp. plus attachments.

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.

Munro, J. L. 1989. Fisheries for giant clams (Tridacinidae: Bivalvia) and prospects for stock enhancement. In: J. F. Caddy, editor. Marine invertebrate fisheries: their assessment and management. New York: John Wiley & Sons. pp. 541-558.

Murawski, S. A. & F. M. Serchuk. 1989. Mechanized shellfish harvesting and its management: the offshore clam fishery of the eastern United States. In: J. F. Caddy, editor. Marine invertebrate fisheries: their assessment and management. New York: John Wiley & Sons. pp. 479-506.

Myers, R. A. & N. J. Barrowman. 1996. Is fish recruitment related to spawner abundance? Fish Bull. 94:707-724.

Olafsson, E. B., C. H. Peterson & W. G. Ambrose, Jr. 1994. Does recruitment limitation structure populations and communities of macro-invertebrates in soft marine sediments: the relative significance o f pre-and post- settlement processes. Oceanogr. Mar. Biol. Ann. Rev. 32:65-109.

O'Malley, M. 1987. Unpublished survey of hard clams in Indian River Bay, Delaware. USEPA.

Peterson, C. H. 2002. Recruitment overfishing in a bivalve mollusk fishery: hard clams (Mercenaria mercenaria) in North Carolina. Can. J. Fish. Aquat. Sci. 59:96-104.

Powell, E. N., E. A. Bochenek, J. M. Klinck & E. E. Hofmann. 2004. Influence of short-term variations in food on survival of Crassostsrea gigas larvae: a modeling study. J. Mar. Res. 62:117-152.

Powell, E. N., J. M. Klinck, E. E. Hofmann, E. A. Wilson-Ormond & M. S. Ellis. 1995. Modeling oyster populations. V. Declining phytoplankton stocks and the population dynamics of American oyster Crassostrea virginica populations. Fish. Res. 24:199-222.

Powell, E. N., M. E. White, E. A. Wilson & S. M. Ray. 1987. Small-scale spatial distribution of a pyramidellid snail ectoparasitic, Boonea impressa, in relation to its host, Crassostrea virginica, on oyster reefs. Mar. Ecol. 8:107-130.

Rice, M. A., C. Hickox & I. Zehra. 1989. Effects of intensive fishing effort on the population of quahogs, Mercenaria mercenaria (Linnaeus 1758) in Narragansett Bay. J. Shellfish Res. 8:345-354.

Ricker, W. E. 1954. Stock and recruitment. J. Fish. Res. Bd. Canada 11:559-623.

Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. 119, Fish. Res. Bd. Canada. 382 pp.

Tarnowski, M. Personal Communication. Maryland Department of Natural Resources, Annapolis, MD.

Thomas, F. I. M. 1994. Transport and mixing of gametes in three free-spawning polychaete annelids, Phragmatopoma californica (Feweks), Sabellaria cementarium (Moore) and Schizobranchia insignis (Bush). J. Exp. Mar. Biol. Ecol. 179:11-27.

Wahle, R. A. 2003. Revealing stock-recruitment relationships in lobsters and crabs: Is experimental ecology the key? Fish. Res. 65:3-32.

Young, D. K. & M. W. Young. 1978. Regulation of species density of seagrass-associated macrobenthos: evidence from field experiments in the Indian River estuary, Florida. J. Mar. Res. 36:569-593.

Yund, P. O. 2000. How severe is sperm limitation in natural populations of maren free-spawners? Trends EcoL Evol. 15:10-13.

Yund, P. O. & S. K. Meidel. 2003. Sea urchin spawning in benthic boundary layers: Are eggs fertilized before advecting away from females? Limnol. Oceanogr. 48:795-801.

Zheng, J. & G. H. Kruse. 2003. Stock-recruitment relationships for three major Alaskan crab stocks. Fish. Res. 65:103-121.

JOHN N. KRAEUTER, (1) STUART BUCKNER (2) AND ERIC N. POWELL (1)

(1) Haskin Shellfish Research Laboratory, Institute of Marine and Coastal Science, Rutgers University, Port Norris, New Jersey; (2) Town of Islip, DEC, Islip, New York

TABLE 1.
Linear, Log, 2nd order Polynomial and Power function models for hard
clam populations in Islip waters of Great South Bay, NY from 1978
to the data year with correlation coefficients for each model.

Data
Year         Linear                         [R.sup.2]

1999   y = 0.097x-0.1371                      0.36
2000   y = 0.1061x-0.1052                     0.40
2001   y = 0.1123x-0.0841                     0.45
2002   y = 0.1186x-0.0628                     0.48
2003   y = 0.1237x-0.0453                     0.51

Data
Year           Log                          [R.sup.2]

1999   y = 0.3289Ln(x)+0.0958                 0.46
2000   y = 0.347Ln(x)+0.0756                  0.52
2001   y = 0.3427Ln(x)+0.0801                 0.57
2002   y = 0.346Ln(x)+0.0766                  0.61
2003   y = 0.3508Ln(x)+0.0714                 0.64

Data
Year               Polynomial               [R.sup.2]

1999   y = 0.0348[x.sup.2]+0.3428x-0.2307     0.49
2000   y = 0.0371[x.sup.2]+0.3619x-0.265      0.55
2001   y = 0.0358[x.sup.2]+0.3511x-0.2464     0.59
2002   y = 0.036[x.sup.2]+0.3531x-0.2497      0.63
2003   y = 0.0365[x.sup.2]+0.3572x-0.2571     0.66

Data
Year             Power                      [R.sup.2]

1999   y = 0.1136[x.sup.1.1949]               0.48
2000   y = 0.0989[x.sup.1.3201]               0.55
2001   y = 0.0976[x.sup.1.3321]               0.61
2002   y = 0.0897[x.sup.1.4117]               0.66
2003   y = 0.084[x.sup.1.4735]                0.70

TABLE 2.
Approximate adult density [m.sup.-2] at which various spawner
recruit curves indicate insufficient adults to provide for a
sustainable population.

Year   Linear   Natural Log   Polynomial   Power

1999    1.41       0.75          0.73      0.001
2000    0.99       0.80          0.80      0.003
2001    0.75       0.79          0.76      0.003
2002    0.53       0.80          0.77      0.005
2003    0.37       0.82          0.78      0.006

TABLE 3.
Minimal and maximal adult population density (number [m.sup.-2])
according to a polynomial model for Great South Bay, NY, and the
density at which hard clam recruitment reaches maximum.

             Maximum Recruits        Adult Density of 0 Recruits

         Recruits        Adults       Low Adults    High Adults
Year   # [m.sup.-2]   # [m.sup.-2]   # [m.sup.-2]   # [m.sup.-2]

1999       0.61           4.96           0.73           9.10
2000       0.62           4.92           0.80           8.96
2001       0.61           4.94           0.76           9.05
2002       0.61           4.95           0.77           9.04
2003       0.62           4.93           0.78           9.00

TABLE 4.
Some bay wide estimates of hard clam population densities.

                              Number
Location                    [m.sup.-2]   Source

Narragansett Bay, RI          2.50       Lazar, Ganz and Valliere
                                           (1999)
Narragansett Bay, RI          3.14 *     Lazar, Ganz and Valliere
                                           (1999)
Islip portion of GSB, NY    6.18-0.86    Buckner unpublished
Raritan Bay, NJ 1963          5.06       McCloy and Joseph (1985)
Raritan Bay, NJ 1983          3.22       McCloy and Joseph (1985)
Raritan Bay, NJ 1999          7.72       Celestino (2000)
Navesink River, NJ            4.81       McCloy and Joseph (1983)
Shrewsbury River, NJ          4.16       McCloy and Joseph (1983)
Metedeconk River, NJ          0.50       McCloy and Joseph (1985)
Great Bay, NJ                 1.33       Joseph (1989)
Shark River, NJ               3.10       McCloy and Joseph (1985)
Manasquan River, NJ           3.80       McCloy and Joseph (1985)
Barnegat Bay, NJ              1.43       Joseph (1986)
Little Egg Harbor, NJ         3.98       Joseph (1987)
Rehoboth Bay, DE 1967         1.71 **    Humphries and Daiber
                                           (1968)
Rehoboth Bay, DE 1975-76      3.45       Cole and Spence (1977)
Indian River Bay, DE 1967     1.83 **    Humphries and Daiber
                                           (1968)
Indian River Bay, DE          2.06       Cole and Spence (1977)
  1975-76
Indian River Bay, DE 1987     2.23       O'Malley (1987)
Chincoteague Bay, MD 1952     1.34       Tarnowski, Personal
                                           Communication
Chincoteague Bay, MD          1.11 **    Wells (1957)
Chincoteague Bay, MD 1969     1.09       Tarnowski, Personal
                                           Communication
Chincoteague Bay, MD 1993     0.27       Tarnowski, Personal
                                           Communication
Chincoteague Bay, MD 2003     0.27       Tarnowski, Personal
                                           Communication
Isle of Wight Bay, MD 1953    1.19       Tarnowski, Personal
                                           Communication
Isle of White Bay, MD 2003    0.28       Tarnowski, Personal
                                           Communication
Assawoman Bay, MD 1953        1.00       Tarnowski, Personal
                                           Communication
Assawoman Bay, MD 2003        0.18       Tarnowski, Personal
                                           Communication
Sinepuxent Bay, MD 1953       1.04       Tarnowski, Personal
                                           Communication
Sinepuxent Bay, MD 2003       0.23       Tarnowski, Personal
                                           Communication

* Adjusted data for a 62.5% dredge efficiency.

** Tongs and rakes.