Early Post-Settlement Growth in Wild Eastern Oyster (Crassostrea virginica Gemlin 1791) Populations
Management and restoration of wild oyster populations with the ecosystem services they provide require detailed understanding of oyster population dynamics, including temporally and spatially varying growth. Much of the existing literature documenting growth rates for eastern oysters (Crassostrea virginica) reports growth for large, protected, and/or hatchery-spawned oysters. By following growth of wild oysters set on planted clamshells in Delaware Bay, we document early growth (within the first year) of 21 wild oyster cohorts settling over 8 years and assess the importance of interannual variability in temperature and salinity. In general, oysters follow a linear growth trajectory in the first year of life, interspersed by periods of little to no growth in the colder months. Wild oysters settling in the Delaware Bay mid-salinity region reach a size between 27 and 33 mm in their first year and tend to reach greater shell heights at 1 year of age in higher salinity years and at temperatures averaging 23 °C. Multi-year, population-level estimates of wild growth such as these are important for understanding changes in restored and managed oyster populations, and resulting ecosystem services, under naturally variable conditions.
KeywordsCrassostrea virginica Early life history Post-metamorphic growth Spat
Oysters are a keystone ecological species. A number of direct and indirect ecosystem services are generated by the oysters themselves such as water filtration (Fulford et al. 2007; zu Ermgassen et al. 2013), benthic-pelagic coupling (Newell 2004), nutrient cycling (Fulford et al. 2010; Kellogg et al. 2013), and via their provision of hard substrate, including juvenile fish habitat (Harding and Mann 2001; Lehnert and Allen 2002; Peterson et al. 2003; Taylor and Bushek 2008), sediment stabilization (Piazza et al. 2005; Scyphers et al. 2011), and wave energy dissipation (Pinsky et al. 2013). Management of wild oyster populations with the ecosystem services they provide requires an understanding of their dynamics which, in turn, requires a basic understanding of oyster demography and growth in response to temporally and spatially varying environmental conditions (Casas et al. 2015).
The overall ecological impact of an oyster population can be modulated by abundance and size structure of the population because these population characteristics are directly linked to the ecological services provided. For example, filtration rate and shell size change ontogenically (Winter 1978; Harding 2007), and a greater overall biomass is able to filter a greater volume of water and provide more shell habitat. Thus, the dynamics of size, age, and abundance of a population determines the ecological impact of that population in the ecosystem. As a result, understanding growth is important for linking population dynamics with the ecological role of bivalve populations.
In addition to the ecological importance of the eastern oyster (Crassostrea virginica), the US fishery for this species is economically important. In 2010, an estimated 18.2 million pounds (8.26 million kg) of meats, valued at US $76.2 million, were landed (Lowther 2011). In Delaware Bay, the commercial oyster harvest is primarily conducted via dredge fishery with recently stable landings of between approximately 75,000 to 90,000 bushels annually (1 NJ bushel = 35 L and yields about five pounds [2.27 kg] of meat) (Ashton-Alcox et al. 2015). Like many, C. virginica stocks along the Atlantic and Gulf coasts of the USA; Delaware Bay oyster stocks suffer large disease losses, adding to the complexity of management in the species. Clear definition of early growth is important for identification of year classes for use in fishery and population management models (Hilborn and Walters 1992) and for linking population dynamics with the ecological role of that population (Andersen et al. 2013).
Extensive literature exists documenting growth rates for eastern oysters; however, many of these growth studies were performed with large (>50 mm) individual (unattached) oysters in trays or other containers (Kraeuter et al. 2007). Trays can cause growth rates to be skewed high because (1) they provide protection from predation, which can lead to higher growth rates in oysters (Walne and Davies 1977), and (2) growth in trays is faster than that on bottom (Kraeuter et al. 2007). Likewise, the use of larger oysters misses growth during the first year immediately following settlement, a critical stage in sessile benthic species that controls subsequent population dynamics (Keough and Downes 1982; Hunt and Scheibling 1997). Growth rates of these early size classes can vary annually and spatially and are influenced by a number of factors such as temperature, salinity, and food availability (Shumway 1996). The salinity gradient present in estuaries creates a corresponding gradient in oyster growth such that oysters in the upper estuary areas (low salinity) typically grow slower than those in the higher salinity regions of an estuary (Kraeuter et al. 2007). Interannual variability in environmental conditions such as temperature and salinity likely influences early growth rates similarly.
The purpose of this study is to examine early post-settlement growth rates for wild eastern oysters and test how settlement timing, temperature, and salinity influence early growth. The extent of the spatial and temporal replication in this study allowed a unique and robust characterization of the interannual and environmental variability on early oyster growth under natural, unmanipulated conditions.
Sampling Spat and Juveniles
Sampling at each shell plant site began approximately 8 weeks following planting and continued monthly during the growing seasons (May to September) for 2 years after planting or until it was difficult to distinguish the first cohort to settle on the planted shell from subsequently settled oysters. Samples were taken by dredge with a maximum of five tows at each plant site each month. Shell height (longest dimension from hinge to growing edge) was measured for a maximum of 100 randomly selected live oysters attached to planted clamshell at each site each month. Only yearlings and spat attached to planted shell were measured; oysters that set on any substrate other than planted shell were excluded from the survey. Cohorts with fewer than 20 individuals (observations) per sample were excluded from further analysis.
Calculation of Growth
Oyster growth was calculated for the first cohort of oysters that settled at each of the shell plants. We calculated early growth in two ways. The first was to estimate the shell height of each cohort at 1 year of age (365 days since settlement). When a measurement was not available exactly on day 365, the average shell height of a cohort was interpolated linearly between the two average cohort height measurements bounding day 365. The second growth estimate was a calculation of the time required for each cohort to grow to 20 mm, the size defined as the transition from “spat” to “juvenile oyster” for purposes of the Delaware Bay annual oyster stock assessment (Aston-Alcox et al. 2015). The number of days required for each cohort to attain an average shell height of 20 mm was estimated by linear interpolation between the two observations of average cohort shell height that bounded 20 mm. For example, if the average shell height was 15 mm on day 40 and 25 mm on day 70, it would be assumed that the cohort reached 20 mm at day 55. Finally, each of the cohorts was characterized as early or late settlers. If the cohort reached 20 mm before day 200, it was considered an early settling cohort; if it reached 20 mm on or after day 200, it was considered a late settling cohort. This distinction was necessary because cohorts setting later in the year are smaller (<20 mm) at the time that they enter the period of almost no growth through their first winter resulting in a slower initial growth rate.
Temperature and Salinity
Bottom water temperature and salinity at each sample site was recorded using a YSI® Model 30 handheld meter during each sampling event. For each cohort, temperature and salinity measurements were averaged over the monthly measurements made, while the cohort was tracked during growing seasons (May to September).
Average shell size was calculated for each cohort, and differences in calculated growth metrics (days to reach 20 mm and shell height at 1 year) between early and late settling cohorts were compared using analysis of variance (ANOVA). The relationship between environmental variables (average temperature and salinity) and each of the calculated growth metrics was evaluated using generalized linear modeling (GLM, R Development Team 2007) to find the best fit model using Akaike’s information criterion (AIC; Akaike 1974).
Summary information about each shell plant used in this study
This indicates that the early growth of a given cohort is governed by a number of factors including location within the bay (bed), environmental conditions of a given year (temperature and salinity), and the timing of settlement, as well as interactions among those factors.
Monitoring growth rates of wild spat is difficult due to the inability to identify when settlement occurs and difficulties in marking and tracking specific cohorts. Here, we used a unique approach for tracking the early growth of oyster cohorts under a variety of environmental condition over a number of years. By following the size of oysters settling on planted clamshells, we were able to monitor and compare growth of 21 oyster cohorts over the first 2 years after settlement and assess the importance of interannual variability in temperature and salinity in early growth. In general, oysters in the Delaware Bay follow a relatively linear growth trajectory in the first year of life, interspersed by periods of little to no growth in the colder months, and tend to reach greater heights at 1 year in higher average salinity and at temperatures averaging 23 °C. Faster growing cohorts will achieve greater biomass and shell surface area in a shorter period of time, thereby providing more filtration capacity and shell habitat in a shorter period of time relative to slower growing cohorts, all else being equal (i.e., abundance and mortality). The factors that were important in determining early growth rates included timing of settlement, temperature, salinity, and location within the bay. Success of oyster restoration efforts and of sustainable management of wild oyster populations requires a basic understanding of variability in early growth and the role of factors that influence it (Paynter et al. 2010; Casas et al. 2015).
Growth differences on the individual level can have a pronounced effect on the description of the growth of a group, and ignoring this individual variability can result in biased estimates of the mean parameter values. The effect of individual variability is particularly important when interpreting growth increment data and can result in overestimates of the mean height at age of a cohort (Sainsbury 1980). Nonetheless, on a population level, early growth in oysters appears to follow a linear pattern rather than a more commonly used saturation function such as a von Bertalanffy model (Fabens 1965). Lester et al. (2004) argue that fish exhibit differences in somatic growth before and after maturity with prereproductive growth best modeled as a linear function and a von Bertalanffy function more appropriate after maturity. Our data suggests that oysters may also show this pattern in prereproductive growth.
For the purposes of fishery assessment of the oyster stock in the Delaware Bay, oysters are considered spat, or within their first year, if they are less than 20 mm shell height (Ashton-Alcox et al. 2015). Those that measure greater than 20 mm are no longer considered yearlings. To test the validity of this assumption, a calculation was made for each cohort to determine the average amount of time it took recently settled cohorts of oysters to grow to 20 mm under a variety of conditions, over a number of years, throughout the Delaware Bay. On average, it took the cohorts 243 days to reach 20 mm, placing the average growth to 20 mm within the first year of life. However, the cohorts varied in the amount of time to grow to 20 mm based largely on whether they set early in the season or late. Early settling cohorts reached the spat transition size in 112 days, while the late settling cohorts required more than twice as many days (309) to reach the same average size. This discrepancy occurs because the late settling cohorts enter into the winter period of low growth before they reach 20 mm, and therefore, the winter adds greatly to the time it takes them to reach the spat transition. This has important implications for how young of the year (spat) are interpreted based on height measurements within the first year. It is worth noting here that for any given year in our dataset, cohorts either settled early or late, meaning there are no years during which both categories of settlement timing occur. This limits our ability to discern the influence of settlement timing while holding interannual conditions constant; however, observations from settlement groups distribute equally across the annually varying range of environmental conditions suggesting sufficient coverage of conditions and cohort types.
Despite the vast difference in the time it takes an early settling or late settling cohort to reach 20 mm, the difference in the average size at 1 year of age among these two groups diminishes to about 15%. In the cohorts we observed, late settling cohorts were nearly able to catch up to the size of the early settling cohorts by 1 year of age. Compensatory growth in oysters has been documented in laboratory and hatchery settings, but in those cases, compensatory growth has largely been attributed to release from competition when cohorts are size sorted (Collet et al. 1999). The contrast between early and late settlers in time to 20 mm (>200% difference) and size at 1 year (∼15% difference) observed in our study could result from a reduction in growth of the early settlers due to spawning within their first year. This reproductive effort would require energy to be diverted from growth to reproduction, thereby slowing average growth of the early settlers who had an initial size advantage and resulting in no size difference between early and late settlers at 1 year of age. Details of spawning and reproductive condition of the animals in this study were not assessed, so this possibility cannot be addressed herein.
An important consideration is that the growth rates estimated here use population-based size frequency to follow growth of a cohort. Inaccuracies in growth rate estimates could come from immigration, emigration, and size-dependent mortality. Oysters are sedentary and attached to hard surfaces; therefore, immigration and emigration are negligible. Additionally, in our assessment, we only include oysters that were attached to planted shell material to prevent inclusion of oysters that might have been moved from other locations as those could be attached to a variety or more ubiquitous materials such as oyster shell and rocks. Nonetheless, size-dependent mortality remains an issue for the interpretation of these data. If larger size classes are preferentially lost from a population between sampling events (due to predation for example), estimated growth would be biased low. Likewise, if new small individuals are added to the population between sampling events, growth would be skewed low. To reduce this bias, only those oysters attached to material that was placed in the environment at a known time were measured to prevent us from mixing older and younger settling cohorts in with the target cohort.
Density dependence (e.g., allee effects, competition for food) or food variability may also affect the growth of early settlers. In general, the abundance of oysters baywide is well below the historic high in the long-term time series suggesting that the system is below carrying capacity (Powell et al. 2009). We might expect density-dependent growth effects to manifest in a population at carrying capacity but are unlikely to observe density dependence in a population well below carrying capacity unless allee effects are operating. We did not observe any pattern in early growth relative to adult or spat density, suggesting that density dependence is not a factor for the cohorts we observed (see Munroe (2016) for a similar example in intertidal clams). Although we did not collect data on food resources directly, a separate study (Powell et al. 2012) looked at food at these and a number of other sites in the bay and found that food well exceeded oyster demands. The Delaware Bay estuary system is highly turbid, and the oyster beds observed in this study lie within the region of the estuarine turbidity maximum (McSweeney et al. 2016) with associated high levels of bottom water particulate organic matter and chlorophyll (Hermes and Sikes 2016), suggesting that these oysters may not experience food limitation.
The data provided here, representing broad spatial and temporal replication, allows a robust characterization of the interannual and environmental variability on early oyster growth under natural, unmanipulated conditions. The rates we derive demonstrate that under most circumstances, wild oysters settling in Delaware Bay reach a size between 27 and 33 mm in their first year and that growth is positively influenced by annual salinity conditions. Although these rates are lower than other published rates, the rates may be a reflection of cooler, more northern conditions in Delaware Bay and may be a more realistic reflection of wild population-level growth rates of relevance to restoration and fishery management strategies. Multi-year estimates of wild growth such as these are important for understanding population-level changes in oysters and resulting ecosystem services under naturally variable conditions.
We gratefully acknowledge the longstanding cooperation with the New Jersey Department of Environmental Protection and the Delaware Bay Section of the New Jersey Shellfisheries Council, with whom shell-planting efforts are conducted each year. E. McGurk, I. Burt, and J. Gius were integral in data collection and shell plant monitoring. Support for shell planting and monitoring was provided by the state of New Jersey in consultation with the Delaware Bay Section of the Shellfisheries Council and Section 1135 of the USACE Continuing Authorities Program. This work was partially supported by the USDA National Institute of Food and Agriculture Hatch project accession numbers 1002345 and 1009201 through the New Jersey Agricultural Experiment Station, Hatch projects NJ32115 and NJ32114.
- Ashleigh, Lowther 2011. Fisheries of the United States 2010. National Oceanic and Atmospheric Administration, National Marine Fisheries Service, Current Fishery Statistics No. 2010. Silver Springs, Maryland, USA. 104 pp.Google Scholar
- Ashton-Alcox, K., Bushek, D., Morson, J., and Munroe, D. 2015. Report of the 2015 Stock Assessment Workshop: New Jersey Delaware Bay Oyster Beds (17th SAW) February 10–11, 2015. 121 pp.Google Scholar
- Babb, R.M., J. Hearon, C. Tomlin, D. Bushek, K. Ashton-Alcox, and E. Powell. 2008. The Delaware Bay oyster (Crassostrea virginica) restoration program. Journal of Shellfish Research 27(4): 987.Google Scholar
- Fabens, A.J. 1965. Properties and fitting of the von Bertalanffy growth curve. Growth 29: 265–289.Google Scholar
- Harding, J.A., and R. Mann. 2001. Oyster reefs as fish habitat: opportunistic use of restored reefs by transient fishes. Journal of Shellfish Research 20: 951–959.Google Scholar
- Hofstetter, R.P. 1963. Study of oyster growth and population structure of the public reefs in East Bay, Galveston Bay and Trinity Bay. Texas Game and Fish Commission. Project No. MO-R-4 Sept 1961–1962. 23 pp.Google Scholar
- Mann, R., and D.A. Evans. 2004. Site selection for oyster habitat rehabilitation in the Virginia portion of the Chesapeake Bay: a commentary. Journal of Shellfish Research 23: 41–49.Google Scholar
- McSweeney, J.M., R.J. Chant, and C.K. Sommerfield. 2016. Lateral variability of sediment transport in the Delaware estuary. Journal of Geophysical Research: Oceans 121: 725–744.Google Scholar
- Menzel, R.W., and S.H. Hopkins. 1951. Report on experiments to test the effects of oil well brine or ‘bleedwater’ on oysters at Lake Barre oil field. Vol. 1. 1–130. Report to Texas A. & M. Research Foundation Project Nine. June 26, 1951Google Scholar
- Menzel, R.W. 1955. Some phases of the biology of Ostrea equestris Say and a comparison with Crassostrea virginica (Gmelin). Publications of the Institute of Marine Science 4: 69–153.Google Scholar
- Moore, H.F. 1899. Report on the oyster beds of Louisiana. Report US Commission of Fisheries. 24: 45–100.Google Scholar
- Newell, R.I. 2004. Ecosystem influences of natural and cultivated populations of suspension-feeding bivalve molluscs: a review. Journal of Shellfish Research 23: 51–62.Google Scholar
- Powell, E.N., J.M. Klinck, K.A. Ashton-Alcox, and J.N. Kraeuter. 2009. Multiple stable reference points in oyster populations: biological relationships for the eastern oyster (Crassostrea virginica) in Delaware Bay. Fisheries Bulletin 107: 109–132.Google Scholar
- R Development Core Team. 2007. R: A language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. ISBN3–900051–07-0, URL: http://www.R-project.org.
- Shaw, W.N. 1962. Raft culture of oysters in Massachusetts. Fishery Bulletin 61: 481–495.Google Scholar
- Shaw, W.N. 1963. Comparison of growth of four strains of oysters raised in Taylors pond, Chatham, mass. Fishery Bulletin 63: 11–17.Google Scholar
- Shumway, S.E. 1996. Natural environmental factors. In The eastern oyster Crassostrea virginica, ed. V.S. Kennedy, R.I.E. Newell, and A.F. Eble, 467–513. College Park: Maryland Sea Grant College.Google Scholar