1
Introduction
FRAMING THE ISSUE
As the total worldwide fisheries yield from exploitation of wild stocks has declined, the production from mariculture, defined as the cultivation of organisms in their natural marine environment, has increased (Food and Agriculture Organization of the United Nations, 2009). This pattern is especially evident for bivalve molluscs, which are the focus of this report. The recognition that, even in developed countries with professional fisheries managers, wild-stock fish, shellfish, and bivalve molluscs have not always been sustainably harvested (e.g., Jackson et al., 2001a; Lotze et al., 2006) leads to concerns over how coastal policies can facilitate expanding mariculture to meet rising demand while management is conducted in a way to preserve ecosystem integrity and sustainability. Relative to the global pattern, the growth of bivalve mariculture has lagged in the United States. Consequently, there may be the opportunity and perhaps growing incentives for growth in this sector of the fishing industry in the United States, making a review of best management practices (BMPs) for sustainability a timely contribution.
The development of living natural resource management has followed a progression from its virtual absence, when the intensity of exploitation was low, to an approach based upon attempts to model and limit harvest of individual species stocks to levels that are sustainable. Then more recently, resource management has evolved to an ideal of sustainability of the integrity of the broader ecosystem responsible for producing the targeted stocks (Food and Agriculture Organization of the United Nations,
2008a). An analogous progression in molluscan mariculture management approaches appears to be developing (Secretariat of the Convention on Biological Diversity, 2004). Critical questions regarding impacts of bivalve mariculture on the natural ecosystem need to be addressed in order to preserve natural populations of fish and wildlife and to sustain ecosystem services of the ocean. In brief, molluscan mariculture can be included within comprehensive, spatially explicit, ecosystem-based management (EBM) of ocean and estuarine systems. Despite broad consensus for development of EBM of the oceans (Pew Oceans Commission, 2003; U.S. Commission on Ocean Policy, 2004) and seminal conceptual characterizations of the principles to be included in EBM (e.g., Grumbine, 1994; Christensen et al., 2006), practical implementation of EBM, especially for the oceans, has been slow and difficult (Arkema et al., 2006).
A recent workshop report by the Food and Agriculture Organization of the United Nations (2008b) includes contributions from many experts to answer the question of how an EBM scheme for aquaculture could be developed to preserve and sustain natural ecosystem integrity. This committee used the concepts in this workshop report to make additional progress in identifying the issues involved in achieving sustainable mariculture of suspension-feeding bivalves. This committee’s report was written to provide a blueprint for development of EBM for molluscan mariculture. As such, it was prepared in response to the committee’s statement of task to “develop recommendations for BMPs for shellfish [i.e., bivalve molluscan] mariculture to maintain ecosystem integrity.” Several specific questions were included in the complete task to the committee (Appendix A), the answers to which required inclusion of the following contributions. The committee conducted a review to characterize the various types of bivalve mariculture operations and the processes through which they have potential to affect the structure and function of the natural ecosystem. The uncertainties associated with these potential ecosystem impacts are identified, along with suggestions on research needs that could help reduce uncertainty and lead toward development and implementation of spatially explicit ecosystem-based mariculture planning that could enhance benefits and minimize negative impacts. Such an approach required consideration of the ways in which molluscan mariculture and wild-stock fisheries are related. Because cultured molluscs often include nonnative species, this report explicitly addresses the risks and the BMPs and performance standards associated with nonnative bivalve culture. The discussion of BMPs and standards, and the subsequent findings and recommendations, are intended for regulators, resource managers, and the mariculture industry. In addition, this report provides a framework for socioeconomic assessment of bivalve mariculture, thereby acknowledging that humans are an integral part of the ecosystem and that food
production from the ocean plays an important role in its use and represents one valuable ecosystem service.
COMMUNITY STRUCTURE AND FOOD WEBS
Ecologists study ecosystems because exploring the interplay between the physicochemical environment and living organisms is critical to developing a holistic understanding of the organizational processes that control species abundances and dynamics. Yet, ecosystems seldom, if ever, have discrete boundaries. Even large lakes can trickle into adjacent systems and exchange nutrients with terrestrial ecosystems (Power, 2001). To complicate the analytical challenge further, plant and animal populations vary in space and through time. Among the types of temporal variation, most critical to management is long-term change driven by human intervention. Few had recognized or acknowledged the phenomenon of shifting ecological baselines until high-profile reports alerted scientists, resource and environmental managers, and the public (Pauly, 1995; Jackson et al., 2001a; Lotze et al., 2006) because humans have modified ecosystems progressively over many generations. Our perceptions of what is natural are typically based on our own recollection of the ecosystem state in the past and thus fail to reflect the long history of human intervention that shift those baselines over time, often effectively disguising the pristine state (Pauly, 1995; see Box 1.1). Reconstruction of the past is made especially challenging in the estuaries, lagoons, coastal bays, and shallow coves favored for mariculture because multiple human interventions have combined to move these ecosystems away from their post-Pleistocene glaciation state. This caveat is necessary because humans have been exploiting marine invertebrates for millennia—in Mexico, for example, shell deposits from the Chantuto people created mounds as high as 11 m (Rick and Erlandson, 2009). Centuries of exploitation and pollution continue to influence the individual resident species directly and also the food web in which they are imbedded.
The conceptual basis of ecological understanding and prediction of how organisms affect one another, directly by consumption or indirectly by consuming interacting species one or more steps away, is the food web (broadened to be an interaction web so as to include impacts beyond those of consumption). Elton (1927) introduced the concept of food webs (“food cycles” in his terminology), Tansley (1935) invented and applied the term “ecosystem,” and Paine (1980) discriminated among types of food webs, developing a “taxonomy” that has withstood the test of time. Of Paine’s three basic web categories, descriptive food webs and energy flow webs are essentially illustrative of structure. Only interaction webs focus on how particular linkages within the web’s topology drive or
|
BOX 1.1 Shellfish in Drakes Estero The committee’s first report, Shellfish Mariculture in Drakes Estero, Point Reyes National Seashore, California, discusses the historic ecological baseline of estuarine and lagoonal ecosystems in terms of ecosystem services provided by oysters. Oysters are one of several species of bivalve whose feeding activity maintains water clarity by filtering suspended materials and transferring organic material to the sediment in the form of feces and pseudofeces, a process known as biodeposition. In addition to oysters, other suspension-feeding bivalve species historically harvested for food in California—such as the Pacific gaper clam (Tresus nuttalli), cockle (Clinocardium nuttalli), littleneck clam (Protothaca staminea), and butter or Washington clam (Saxidomus nuttalli)—contribute similar biogeochemical functional attributes. Although there are considerable uncertainties about the historical abundance of native Olympia oysters in Drakes Estero, the cultured non-native oysters to some extent replicate the biogeochemical functions of several species of native bivalves, including Olympia oysters. The habitat services provided by mariculture of Pacific oysters may differ from those provided by native Olympia oysters because the native oyster does not form extensive tall reefs, even where abundant, whereas the rack structures holding strings of cultured Pacific oysters extend about 1 m upwards from the bottom and provide hard-substrate habitat for a nonnative tunicate that covers a substantial portion of the rack and oyster surfaces. The biogeochemical effects of the cultured nonnative oysters, as distinguished from the impacts of mariculturists’ activities, are likely to be small as long as the level of production is low relative to the ecological carrying capacity of the ecosystem. However, this is only one of the issues to be evaluated in developing appropriate management practices for a bivalve mariculture operation and determining whether a site is appropriate. For Drakes Estero, the committee’s first report notes several other considerations, such as the potential for mariculturists’ disturbance of harbor seals and water fowl, as well as policy constraints in an area congressionally designated as Potential Wilderness. |
modify changes in species populations and thereby create community patterns (see National Research Council, 2006).
Some of the underlying interspecific interactions in these webs are grounded in classic observational, comparative studies (e.g., Brooks and Dodson, 1965; Estes and Palmisano, 1974), others in experimental manipulation (e.g., Paine, 1966; Power et al., 1985), and still more in studies that blend these techniques (e.g., Myers et al., 2007). Studies involving interaction webs have generated a vocabulary of their own—top-down organization (predator control of the system’s basic processes) versus bottom-up organization (control by primary productivity)—and have progressed to rejuvenate interest in trophically transmitted (Wootton, 1993; Menge, 1995) and behaviorally mediated (Peacor and Werner, 2001;
Grabowski, 2004) indirect effects. Appreciation of these processes that act on species within a community is necessary in developing the important applied concepts of system carrying capacity and EBM.
The evidence that natural ecosystems, because they are composed of networks of dynamically interacting populations, respond to perturbations comes primarily from experimental manipulations conducted at small spatial scales (e.g., Paine, 1966; Dayton, 1971; Sutherland, 1974; Carpenter and Kitchell, 1992). Menge (1997) asks whether those done in marine intertidal systems were of sufficient duration both to detect the anticipated direct effects but also to generate the indirect ones. Both categories of effects generally appeared simultaneously and became statistically significant within the initial 20–40% of the experimental duration. However, a question more relevant to the expanded spatial scales characteristic of most mariculture operations is whether the interactions defined and demonstrated at smaller scales can be scaled up and also applied at the level of an estuary or even an open ecosystem.
Two primary lessons derived from the outcomes of these experimental studies can be applied to the management of bivalve mariculture. First, many of the resident species are dynamically connected, as reflected in the concept of an interaction web, and these interactions have both direct and indirect consequences. Second, the most effective means of identifying the fundamental dynamics is by intervening in the system. Bivalve mariculture itself is an intervention, as is the addition, or depletion, of a higher, consuming trophic level, or successful establishment (in abundance) of an invasive species. Box 1.2 presents three examples1 to illustrate that species population dynamics are linked and that ecological “surprises” can arise in the form of unanticipated indirect effects of suspension-feeding bivalves.
The denominator common to these three studies is that ecosystems at large spatial scales can have their primary productivity substantially redirected by large populations of suspension-feeding bivalves, which is most clearly demonstrated by the unintended ecosystem intervention of successful establishment and proliferation of a set of nonnative suspension-feeding bivalves. While U.S. molluscan mariculture has not reached the levels provided in the examples above, it is important to understand the potential impacts of high-density culture of both native and nonnative species. Dumbauld et al. (2009) review in detail the nuances
|
BOX 1.2 The Ecosystem Impact of Selected Invasive Bivalves Zebra mussel (Dreissena spp.) This benthic invertebrate is an iconic invasive species that was first identified in North American waters in 1988 and has since spread broadly (Johnson et al., 2006). Zebra mussels are suspension feeders—hence their relevance to the bivalves used in molluscan mariculture—and often occur as dense populations. The community impact of their entry, establishment, and proliferation is best documented from Lake Erie and especially the Hudson River estuary (Strayer et al., 2004; Strayer and Malcom, 2007; Strayer, 2009). They were initially recognized in the Hudson River in May 1991, and 17 months later, their biomass “…exceeded that of all other heterotrophs in the freshwater tidal Hudson…” (Strayer et al., 1999). Measured filtration rates translate to a theoretical turnover of the entire water column in 1.2–3.6 days for this tidally well-mixed estuary. Figure 1.1 identifies the sweeping consequences of this invasion. Phytoplankton and small zooplankton have declined precipitously (-80% and -71%, respectively), as have pelagic fish (-28%), reflecting a dramatic impact on the pelagic food web. On the other hand, the benthic food web has flourished, in large part because of clearer water, enhanced growth of submerged macrophytes, and increased densities of both nearshore invertebrates and fish. Indirect effects are apparent. Bacterial populations have increased, and some bivalve-consuming ducks have benefited, yet some native bivalves may be nearing local extinction, and crayfish are heavily and detrimentally fouled.  FIGURE 1.1 Summary of the effects of the zebra mussel invasion on the Hudson River ecosystem (copyright by the Ecological Society of America; Strayer, 2009). |
|
Corbicula fluminea in the Potomac River This introduced Asiatic bivalve was recognized in the tidal, but basically freshwater, Potomac River in 1977 (Cohen et al., 1984). Prior studies had shown a downstream gradient in phytoplankton abundance. Clam density, about 1,500 per m2, peaked in 1980 and 1981, just prior to a population crash and correlated spatially with a substantial drop in phytoplankton presence along a 16 km stretch of the river. Estimates of the clam filtering rate suggest that 30% of the river’s chlorophyll a along this stretch could be pumped through the C. fluminea in three to four days. The decline in phytoplankton presence appears consistent with the clam’s filtering ability. The bivalves appear to have substantially influenced the nutrient resources common to other suspension feeders. (Indirect effects were not identified.) Potamocorbula amurensis in San Francisco Bay The capacity of suspension-feeding bivalves to alter community structure is clearly illustrated in the largest estuary on the U.S. Pacific coast, the San Francisco Bay estuary. The presence of P. amurensis was recognized by 1986 (Carlton et al., 1990). An individual adult can filter about 4 liters of water per day. That per capita rate, combined with a density of about 16,000 per m2 (Chauvaud et al., 2003), was sufficient to account for the observed suppression of an annual phytoplankton bloom lasting from late spring to fall (Alpine and Cloern, 1992). Associated with this suppression was an approximate 80% decrease in copepod density, presumably due to reduced phytoplankton availability (Nichols et al., 1990). Copepods are a major prey for endangered and federally listed smelt species. The relationship between water rights law, California agriculture, Chinook salmon production, and the Endangered Species Act represents a regulatory quagmire. While bivalve mariculture is not involved, the sweeping consequences to a diverse stakeholder assemblage attributed to a dominant and introduced suspension feeder are obvious. One indirect effect is on diving ducks. These introduced bivalves are readily consumed by Greater Scaup, but because the introduced species has a thicker shell than their normal prey, the per clam food value is reduced, and the duck’s dispersal and over-wintering habits have been altered (Poulton et al., 2002). |
underlying potential shifts in community structure. In general, nutrient availability, the identity of the phytoplankton species, per capita prey growth and predator consumption rates, and respective densities of all major participants are needed at minimum to model the potential for change. Thus, modeling will be an essential component of research to identify the carrying capacity of suspension-feeding bivalves (see Chapter 5), and this is dependent on meeting the challenge of estimating the various rate functions.
Do commercially farmed bivalves, generally capable of filtering a broad size spectrum of prey (even including some invertebrate larvae), influence the local community by simultaneously being a competitor and a predator? All communities in nature are ensembles of dynamically interacting species. Enough detail is now known to be able to predict change following the addition of dense suspension-feeding bivalve populations; however, knowledge is insufficient to predict with confidence the consequences for particular species. The implications for intensive, local development of bivalve mariculture seem obvious—ecosystem impacts can be anticipated although many of them may not be immediately apparent and cannot be predicted with the certainty that stakeholders often demand from resource managers and decision makers. Assessing whether anticipated modifications of the estuarine ecosystem are beneficial or detrimental depends in part on knowledge of historical baselines of bivalve abundance and a synthesis of the net value of direct and indirect impacts. As detailed in the scientific literature and some high-profile reviews (e.g., Lotze et al., 2006), estuaries are largely degraded worldwide. This implies that managing for a return toward historical baselines represents a beneficial change, especially where abundant wild populations of suspension-feeding bivalves played an historical role of providing resilience against eutrophication symptoms (Jackson et al., 2001a).
KEY CONCEPTS
Ecological integrity—the capacity of an ecosystem to support and maintain a balanced, integrated, adaptive community of organisms having an indigenous species composition, diversity, and functional organization comparable to that of similar undisturbed ecosystems in the region (Carignan and Villard, 2002).
Resilience—the capacity of an ecosystem to maintain its characteristic patterns, structures, function, and rates of processes in the face of disturbance or perturbation (Leslie and Kinzig, 2009).
SPECIES CONSIDERED IN THIS REPORT
The bivalve molluscs currently or historically cultured in the United States to market as food include oysters of several species (Crassostrea virginica, C. gigas, C. ariakensis, C. sikamea, Ostrea lurida, and O. edulis), mussels (Mytilus edulis, M. trossulus, and M. galloprovincialis), several venerid (family Veneridae) clams (Mercenaria mercenaria, Protothaca staminea, and Venerupis phillipinarum), scallops (Argopecten irradians), geoducks (Panopea generosa), soft-shell clams (Mya arenaria), cockles (Clinocardium nuttallii), rock scallops (Hinnites giganteus), arks (Anadara transversa and A. ovalis), and razor clams (Siliqua alta, S. costata, and S. patula). Although not a bivalve mollusc but rather a marine gastropod mollusc, this report also discusses abalone (Haliotis spp.) because they are commonly cultured on the U.S. Pacific coast and provide insight into disease issues shared with bivalve molluscs. Culturing of bivalves includes many experimental trials that resulted in failure, such that this list of bivalve types does not imply that bioeconomically feasible culturing has been demonstrated for each group or each species.
These cultured molluscs and the methods of growing them differ in discrete ways. The bivalve species can be subdivided by aquatic environmental regime in which they live (ocean versus estuary versus marine lagoon), relationship to substrate (epifauna on the surface of the hard substrate versus infauna buried within the soft substrate), and major predators (e.g., seaducks for mussels, gastropods for oysters). Likewise, culture techniques fall into clearly distinguishable categories. Culture can be conducted in floating containers, suspended containers or lines, or on the bottom. The jargon of molluscan mariculture also separates the intensity of methods (e.g., use of external inputs, which also translates to density of animals per unit area of culture) from extensive methods (e.g., cast out on the bottom, with or without protective mesh or netting) to semi-intensive (with some external control) to intensive culture within a hatchery where food, aeration, and seawater are provided. Clearly, the method and location of culturing can dictate the kinds and intensities of impacts on other species and the broader ecosystem.
GENERAL APPROACHES TO DEFINING CARRYING CAPACITY
The generic concept of carrying capacity in ecology has been developed to refer to the Malthusian notion that resources in the environment are limited such that no population can grow without limit. Formal definitions of carrying capacity for a population of a given species have emerged from mathematical representations of single-species population growth. The most well known of which is the logistic curve where the rate of population growth approaches exponential as abundance nears zero
and resources do not constrain population growth and then approaches zero as the population size (N) nears some value (K) defined as the carrying capacity of the environment. This simple logistic growth curve is represented as:
where r is the maximum intrinsic rate of natural increase. This logistic growth curve has played and continues to play an important role in formalizing theory in population and community ecology. Species interactions are included by inserting parameters relating to competing species and/or predators and expressing interactions among species through coefficients that reflect the nature and strength of those between-species relationships. The fundamental basis of this type of representation is so deeply engrained in ecology and environmental sciences and has sufficient parallels in economics that the general notion of a single-species carrying capacity is widespread and used in conceptual and mathematical models by resource managers and environmental organizations.
Although the concept of carrying capacity is generally understood, there are numerous alternative bases on which to set the carrying capacity for molluscan mariculture, each with different implications for management. The strict application of the logistic growth curve to stocking of suspension-feeding bivalves would lead to a hydrographically defined water body in which individual bivalve seed would become stunted in growth because of exhaustion of the resources required for growth. This situation would be unacceptable to the growers because when food demands of the bivalves cannot be met and growth is stunted, the culture operations would cease to remain financially viable. Furthermore, the impacts of high-bivalve density can be expressed on several different spatial scales, dependent upon renewal processes and rates for suspended foods. Growth rates of individual suspension-feeding bivalves decline with local density on scales as fine as 1 m2 (Peterson, 1982; Peterson and Black, 1987) and within individual mariculture operations (Newell et al., 1998; Drapeau et al., 2006). These sorts of consequences probably have more relevance to management decisions of the individual grower, but such localized depletion of suspended foods would also affect natural populations of suspension feeders of all sorts, including zooplankton as well as benthic invertebrates.
As alternatives to estimating carrying capacity based on maintaining adequate supplies of suspended foods for the cultured bivalves, carrying capacity can instead be based on sustaining ecosystem needs or on social acceptance of mariculture. Determining the ecosystem needs of other suspension feeders in the water column and on the bottom and assessing
the ecosystem capacity to process the release of dissolved nutrients and the biodeposits of feces and pseudofeces represents a scientific challenge but moves management of bivalve mariculture closer to an EBM ideal of sustaining ecological integrity. There is precedent within the management regimes of some wild-stock bivalve fisheries. For example, mussel production in the Netherlands is allocated to fishermen only after inferred demand from mollusc-eating birds is satisfied (see Wadden Sea box in Chapter 4). Newly developing approaches to this challenge are even producing methods of comparing benefits of fishery production to declines in important species in the ecosystem that are indirectly affected through the food web impacts (Richerson et al., 2009). Setting a carrying capacity for bivalve mariculture sensitive to the resource demands of non-commercial species would lead to a lower carrying capacity than one based on the cultured mollusc production alone. The social carrying capacity reflects the local public attitudes toward use of waters for a variety of alternative, largely incompatible purposes. Because the social carrying capacity is likely to vary widely from place to place determined by multiple user conflicts, in large part dependent upon the rising human population density along the coasts (Diana, 2009), less-populated regions are expected to be more tolerant of bivalve mariculture interventions as a “new” coastal use. Social carrying capacity probably requires some form of a survey tool to determine prevalent local attitudes because of its place-specific nature. Carrying capacity based on social considerations will generally be a lower, often far lower, number of cultured molluscs than the level that could still supply ecosystem needs and thus sustain ecological integrity. In addition, these different types of carrying capacity are not independent. Many stakeholders will provide social input that reflects an environmentalist commitment to sustaining ecological integrity, such that this consideration will contribute to determining social carrying capacity.
REPORT ORGANIZATION
This report is organized to review the challenges, constraints, and benefits of maintaining or restoring ecosystem integrity in the presence of bivalve mariculture. Chapter 2 describes BMPs and performance standards for bivalve mariculture. Chapter 3 identifies the ecological effects of bivalve mariculture. Chapter 4 discusses the relationship between bivalve mariculture and wild-stock harvest. Chapter 5 analyzes carrying capacity as it relates to bivalve mariculture. Chapter 6 focuses on the economic and policy factors affecting bivalve mariculture activities, and finally Chapter 7 provides some concluding synthetic perspectives on ecosystem services of bivalves. Appendix A includes the committee’s verbatim statement of task, and Appendix B presents the committee and staff biographies.
