greening of the blue revolution: efforts toward ...

Moss

Page 1

GREENING OF THE BLUE REVOLUTION: EFFORTS TOWARD ENVIRONMENTALLY RESPONSIBLE SHRIMP CULTURE Shaun M. Moss, Steve M. Arce, Brad J. Argue, Clete A. Otoshi, Fernanda R.O. Calderon, and Albert G.J. Tacon The Oceanic Institute, 41-202 Kalanianaole Hwy., Waimanalo, HI 96795 USA

ABSTRACT Shrimp aquaculture has been criticized by organizations and individuals that charge the industry as being environmentally irresponsible. Publications such as Murky Waters by Goldburg and Triplett (1997) and the recent article in Nature by Naylor et al. (2000) identify a number of environmental problems associated with shrimp culture, including habitat destruction, water pollution, non-native introductions, collection of wild shrimp, and excessive use of marine protein in shrimp feeds. Although some of these charges have merit, others are not supported by scientific data. Clearly, it is in the industry’s best interest to engage in environmentally responsible methods of production if long-term viability of the industry is to be achieved. Commercial farmers and researchers continue to develop and evaluate approaches to shrimp culture that protect both the natural environment as well as the shrimp culture environment. For example, by reducing water exchange rates, effluent discharge is minimized, thereby reducing nutrient and biological pollution in surrounding bays and estuaries. In addition, because influent water can serve as a vector for disease, the potential for pathogen introduction into the shrimp culture environment is reduced. The use of domesticated, specific pathogen free (SPF) shrimp eliminates the need for wild-caught shrimp and reduces the risk of transferring pathogens to noninfected areas. Also, domesticated shrimp can be selectively bred to improve economically important traits such as growth and disease resistance. Vegetable proteins can be used to replace or supplement marine proteins commonly used in shrimp feeds and this may relieve fishing pressures on marine resources, including pelagic fisheries. Vegetable-based diets would have the added benefit of being free of shrimp pathogens and would be less costly than diets containing marine proteins. Researchers at the Oceanic Institute are developing shrimp production technologies that integrate several of the approaches identified above. Specifically, we have cultured SPF, genetically improved Pacific white shrimp (Litopenaeus vannamei) in a recirculating raceway designed for pathogen exclusion and have produced 4 kg of shrimp/m2 using 370 liters of water per kg of shrimp. In addition, we have evaluated vegetable-based diets and have achieved shrimp growth rates greater than 1 g per week. The heritability (h2) estimate for growth when shrimp

were fed vegetable-based diets was high (mean half-sib h2 = 0.40 + 0.30 (+ SE)), and it may be possible to select shrimp for rapid growth on these diets. Importantly, through advanced research, the shrimp farming industry will be able to expand into areas away from the coast with greater control against the spread of disease and without adversely affecting the environment.

INTRODUCTION The shrimp aquaculture industry expanded significantly throughout Asia and Latin America during the 1980’s and this expansion was generated largely by abundant wild seed, static supplies of shrimp from capture fisheries, and high profits from cultured shrimp (Fast and Menasveta 2000). In 1999, farmers produced an estimated 814,250 metric tons of shrimp (Rosenberry 1999), and this represents about 25% of the total shrimp production worldwide. Despite high levels of production, shrimp farmers have suffered significant economic losses in recent years due to disease problems that have plagued the industry. In Asia, mortalities of cultured shrimp due to White Spot Syndrome Virus (WSSV) and Yellow Head Virus (YHV) have resulted in economic losses of about US$1 billion per year since 1994 (Lightner et al. 1998). Similarly, in Ecuador alone, Taura Syndrome Virus (TSV) has been responsible for an estimated US$400 million loss in revenue per year, and this virus has had an equally devastating impact in other shrimp farming countries of the Americas (Brock et al. 1997). In addition to disease problems, the shrimp farming industry is confronted with charges of environmental irresponsibility, including habitat destruction, water pollution, non-native introductions, collection of wild shrimp, and the use of marine protein in shrimp feeds (Goldberg and Triplett 1997, Naylor et al. 1998, Naylor et al. 2000, Table 1). Although there are examples that illustrate the negative environmental impacts of shrimp farming (Chua et al. 1989, Primavera 1991), these impacts often have resulted from poor planning and management rather than something inherently destructive about shrimp culture per se (Boyd 1996, Boyd and Clay 1998). Clearly, it is in the shrimp farming industry’s best interest to

Craig L. Browdy and Darryl E. Jory editors. The New Wave, Proceedings of the Special Session on Sustainable Shrimp Culture, Aquaculture 2001. The World Aquaculture Society, Baton Rouge, LA USA.

Greening of the Blue Revolution

Page 2

Table 1. Potential negative impacts of shrimp farming on the natural environment and shrimp culture environment.

Environmental Concern

Negative Impact on Natural Environment

Habitat Modification

• • •

loss of nursery habitat increase erosion increase vulnerability to storm damage

Water Pollution

• • • •

cause eutrophication stimulate toxic algal blooms reduce property values decline in fisheries

•

• •

alter structure of native aquatic communities change genetic composition of natural populations reduce biodiversity introduction/spread of pathogens

• •

deplete native shrimp populations cause by-catch problems

Non-Native Introductions

•

Collection of Wild Shrimp

Use of Marine Protein in Feeds •

decline in pelagic fisheries

engage in environmentally responsible methods of production. Shrimp farmers and researchers continue to develop and evaluate approaches to shrimp culture that protect both the natural and shrimp culture environments. In this paper, we review several of the major environmental concerns associated with shrimp farming, identify approaches to shrimp culture that mitigate negative environmental impacts, and highlight research conducted at the Oceanic Institute where scientists are developing and evaluating shrimp production technologies that are environmentally responsible.

ENVIRONMENTAL CONCERNS Habitat Modification Shrimp farming activities have been implicated in causing habitat modification of coastal wetlands, especially mangrove destruction (Hatcher et al. 1989, Primavera 1991, Primavera 1995, Naylor et al. 2000). Mangrove habitats are ecologically important for several reasons. In addition to serving as nursery grounds for fish and shellfish (Robertson and Duke 1987, Primavera 1998), they minimize erosion, protect shorelines from storm damage, and are valuable sources of

Negative Impact on Shrimp Culture Environment • • • •

decrease productivity due to acid soils poor pond drainage exposure to erosion and storm damage exposure to disease vectors

• •

self-pollution of ponds introduction/spread of pathogens

•

introduction/spread of pathogens

• •

introduction/spread of pathogens inability to benefit from genetic improvement

• •

introduction/spread of pathogens high cost to shrimp farmer

food, fuel and other products (Lugo and Snedaker 1974, Hatcher et al. 1989). Although there has been a significant worldwide loss of mangrove habitat over the past several decades (Ong 1995, World Resources Institute 1996), the extent to which shrimp farming has contributed to this loss has been debated. According to Naylor et al. (2000), the “loss in wild fisheries stocks due to habitat conversion associated with shrimp farming is large.” Implicit in this statement is the assumption that significant tracts of mangrove and other coastal habitats have been converted into shrimp ponds. However, Boyd (1996) reports that only 6.2% of the world’s mangrove resources have been converted for such use, and the actual percentage may be lower (Macintosh and Phillips 1992). One of the challenges in assessing the impact of shrimp farming on the loss of mangrove habitat is the difficulty in determining if mangrove land was reclaimed for other uses prior to being converted into shrimp ponds. For example, Naylor et al. (2000) indicate that 65,000 hectares of mangroves in Thailand were converted into shrimp ponds between 1961 and 1993. However, because of the survey techniques used to make this estimate, it is difficult to assess whether shrimp ponds were developed from virgin mangrove areas or converted from paddy fields, fish ponds, salt evaporation ponds, or other uses (Menasveta 1997). Importantly, 47% of the total mangrove area present in Thailand

Moss in 1961 was destroyed between 1961 and 1986, prior to a major expansion in the shrimp farming industry (Fast and Menasveta 2000). A similar situation occurred in the Philippines where 338,000 hectares of mangrove land was reclaimed between 1968 and 1988, mostly for activities unrelated to shrimp farming (Boyd 1996). There are compelling reasons for farmers to avoid culturing shrimp in mangrove areas. Mangrove soils typically have a high organic content and often are associated with highly acidic conditions (Simpson et al. 1983, Gaviria et al. 1986). Also, mangrove areas typically do not drain adequately for proper pond management. Importantly, shrimp farms located in mangrove areas are exposed to potential disease vectors (including influent water, birds, crabs, and insects), so there is a strong incentive to avoid these areas to grow shrimp. Most shrimp farmers are now aware of the benefits of mangrove habitat and recognize the limitations and risks associated with this habitat for shrimp culture. In fact, newly constructed shrimp farms in Colombia and Madagascar were sited in areas adjacent to mangrove habitat rather than on reclaimed mangrove land (Bador et al. 1998a, Bador et al. 1998b). Shrimp farms in these areas are protected from storms and erosion by intact mangrove habitat, and farmers can use the mangroves to treat pond effluent (Robertson and Phillips 1995). In Vietnam, the government has promoted shrimp-mangrove integrated farming systems where shrimp are cultured in ponds adjacent to mangrove habitat (Binh et al. 1997). Economic analyses from these farming systems indicate that farmers receive a better economic return when they maintain mangrove habitat, and some shrimp farmers are now engaged in replanting efforts (Chamberlain and Barlow 2000). Although shrimp farming has contributed to the loss of mangrove resources in the past (MacIntosh and Phillips 1992), it is likely that the future contribution of shrimp farming to the worldwide destruction of mangroves will be insignificant relative to other human activities.

Water Pollution Most shrimp are cultured in outdoor, earthen ponds that rely on flow-through water exchange, and the rationale for exchanging water depends on the level of production. In extensive shrimp culture, intake water serves as a medium for the introduction of potential food resources or as a source of shrimp seed (Hopkins et al. 1995c). In more intensive systems, water is exchanged because of the perceived benefits of preventing the accumulation of toxic metabolites in order to maintain acceptable water quality (Hopkins et al. 1993). In addition to the continuous, daily discharge of effluent for management purposes, pond water is released into the environment during drain harvest or after heavy rains (Teichert-Coddington et al. 1999, Boyd 2000).

Page 3 There are concerns that aquaculture effluent may contain high concentrations of organic matter and inorganic nutrients that can adversely affect water quality, particularly in receiving waters where the residence time is relatively long (Ziemann et al. 1992). Although most published accounts of water quality deterioration from aquaculture effluent come from fish culture operations (Gowen and Bradbury 1987), shrimp pond effluent has been implicated in causing negative environmental and economic impacts (Chua et al. 1989, Primavera 1991). The quality and quantity of pond effluent is affected by a number of factors, including weather, pond type, pond management, water exchange rates, feed type and feeding regime, stocking density, and the type and life stage of the target species (Ziemann et al. 1992). The shrimp farming industry is in a position to manipulate many of these factors in order to regulate the volume and content of pond effluent and it may be in their best interest to do so. Pond water quality is affected by the quality of influent water (Teichert-Coddington et al. 2000), and the discharge of organically rich, eutrophic water from shrimp ponds may ultimately reduce farm productivity because of self-pollution (Boyd and Musig 1992, Csavas 1994). In addition to negatively impacting the shrimp culture environment, pond effluent may be harmful to the natural environment. For example, Goldburg and Triplett (1997) blame a decline in sport-fishing productivity and a reduction in residential property values on effluents discharged from shrimp farms in southern Texas. However, the perception that all shrimp pond effluent causes environmental damage is inaccurate, and effluent impact depends largely on the quality of the receiving water. In Hawaii, where coastal shrimp ponds are surrounded by a nutrient-poor ocean, even moderately mesotrophic effluent can exceed the state’s water quality standards that are designed to maintain low ambient concentrations of nutrients, pigments, and suspended materials (Ziemann et al. 1992). In contrast, pond water discharged into eutrophic, organically rich receiving waters may have little or no impact. Robertson and Phillips (1995) compared effluent discharged from an intensive shrimp pond with water from an undisturbed mangrove habitat. Although they found that pond effluent contained ammonia and chlorophyll a concentrations that were one to two orders of magnitude greater than in the mangrove, they also reported that maximum concentrations of nitrate, phosphate, and total suspended solids were greater in the pristine mangrove waters. Hopkins et al. (1993) reported that concentrations of total ammonia nitrogen, nitrite-nitrogen, nitratenitrogen, reactive orthophosphate, suspended solids, and organic solids were not significantly different between water from an intensive shrimp pond (that experienced a 25% water exchange per day) and water from an adjacent estuary. Only in vivo fluorescence was significantly greater in shrimp pond

Page 4


water. Clearly, the impact of organic and inorganic additions to natural aquatic ecosystems is not unequivocally negative. In fact, the discharge of aquaculture effluent may have a beneficial effect on biota inhabiting the receiving waters (Weston 1991), and several commercially important invertebrates thrive in organically rich, eutrophic environments (Moss et al. 1992, Jakob et al. 1993). In addition to serving as a transport medium for organic matter and inorganic nutrients, water can be a vector for shrimp pathogens (Lotz 1997, Moss et al. 1998, Lotz and Lightner 2000). Pathogens can enter and infect a shrimp pond via untreated intake water. Once established, pathogens can spread from pond-to-pond or farm-to-farm via untreated effluent discharged into surrounding canals, bays, and estuaries that serve as a common water source for nearby farms. This potential makes water exchange a risky management option for shrimp farmers unless both influent and effluent streams are disinfected. However, the cost of disinfecting large volumes of water is too expensive for most farmers, so alternative strategies to mitigate the introduction of pathogens need to be identified, evaluated, and implemented. These strategies will likely minimize nutrient and biological pollution as well (see below).

Non-Native Introductions Aquaculture activities have resulted in the accidental introduction of non-native species into the wild (biological pollution), and these introductions have been implicated in altering aquatic community structure and genetic composition of native populations, as well as reducing biodiversity (Beveridge et al. 1994, Goldburg and Triplett 1997, Naylor et al. 2000). Although most published accounts of non-native introductions are from fish farms (Courtenay and Williams 1992, Hansen et al. 1999, Fleming et al. 2000), the accidental release of exotic shrimp into the western North Atlantic Ocean has been documented (Wenner and Knott 1992). However, there is no evidence of any ecological problems associated with the escape of non-native shrimp species (Boyd 1996). Of particular concern with the global shipment of cultured shrimp is the introduction and spread of shrimp pathogens into previously uninfected areas (Naylor et al. 2000). The introduction of Infectious Hypodermal and Hematopoietic Necrosis Virus (IHHNV), Hepatopancreatic Parvo-like Virus (HPV), and Monodon-like Baculovirus (MBV) into areas previously devoid of these viruses resulted from the movement of live shrimp intended for aquaculture use (Lightner et al. 1992). In some cases, viral introductions resulted in catastrophic losses due to disease, whereas in other cases impacts were moderate or insignificant. More recently, WSSV, a virus native to the major shrimp farming regions of Asia, has been confirmed in the Americas, including Ecuador (Rosenberry 1999). This introduction has been especially significant because many com-

mercially important shrimp in the Western Hemisphere are highly susceptible to WSSV infection (Lightner et al. 1998). At this time, it is not clear how WSSV initially was introduced from Asia to the Americas. In addition to the transfer of live shrimp for aquaculture use, other potential vectors of this virus into the Americas include ballast water from ships, imported shrimp used for bait, and processing wastes from imported, frozen commodity shrimp. Frozen shrimp can serve as a vector for exotic shrimp viruses as evidenced by the fact that frozen shrimp imported into the United States from Asia contained viable WSSV and YHV (Nunan et al. 1998).

Collection of Wild Shrimp In most shrimp farming regions of the world, farmers rely on the capture of wild shrimp to stock their ponds. Shrimp are caught as postlarvae and stocked directly for growout or as broodstock that are subsequently spawned in captivity to provide seed. Negative environmental consequences of this activity are not well documented, although there are concerns that harvesting wild shrimp for aquaculture purposes can deplete local shrimp populations (Bashirullah et al. 1989). For example, it has been reported that harvesting wild postlarvae to stock shrimp ponds in coastal Ecuador may have changed the dominant species of shrimp caught by local fishermen (Landesman 1994). In China, there are concerns that the continued use of wild broodstock collected from their natural spawning grounds may negatively impact the shrimp fishery (Xin and Sheng 1992). Although there are conflicting opinions as to whether penaeid shrimp exhibit stock-recruitment relationships (Garcia 1989, Ye 2000), there is evidence that overfishing by commercial shrimp trawlers in Indonesia led to a decline in shrimp biomass, a change in the species composition of the catch, and a decrease in the size of individual shrimp caught (Naamin 1987). In addition to concerns about overharvesting of shrimp for aquaculture purposes, large quantities of incidentally caught organisms (by-catch) may be caught with the target shrimp species during collection of postlarvae (Naylor et al. 2000). By-catch usually is discarded because there is no economic incentive to process, release, or save it. Although the ecological ramifications of this activity are not well established in an aquaculture context, Silas (1987) estimated that 10 kg of by-catch are killed for every 1 kg of Penaeus monodon postlarvae that are harvested by fry collectors in the Sunderbans of West Bengal, India. This by-catch to shrimp ratio is similar to what has been reported for the shrimp trawler industry (Andrew and Pepperell 1992, Moss and van der Wal 1998). The use of wild-caught postlarvae or broodstock by farmers poses a serious risk to the shrimp aquaculture industry, unless shrimp are screened for specifically listed pathogens, because wild shrimp may be carriers of virulent viruses. As

Moss we indicated above, WSSV, YHV, TSV, and IHHNV have caused significant economic losses in the major shrimp farming regions of Asia and the Americas, and these losses can be attributed, in part, to the use of wild, infected shrimp. WSSV has been detected in wild populations of several penaeid species, including Penaeus monodon, P. semisulcatus, Marsupenaeus japonicus, Fenneropenaeus penicillatus, Farfantepenaeus duorarum, and Metapenaeus ensis (Lo et al. 1996, Kuo et al. 1997a, Kuo et al. 1997b, Wang et al. 1997, Baldock 1999). Recently, researchers from the University of Arizona and the Oceanic Institute determined that of 104 wild Litopenaeus vannamei broodstock collected off the coast of Panama, two individuals tested positive for WSSV using a gene probe adapted for a “dot blot” hybridization assay, i.e. dot blot test (unpublished data). However, it is important to note that the dot blot test is able to detect moderate to high levels of infection, so the prevalence of WSSV may have been higher than 2% if shrimp were experiencing low-grade infections. With regard to IHHNV, this virus also has been detected in wild populations of several penaeid species, including L. vannamei, L. stylirostris, L. occidentalis, and Farfantepenaeus californiensis (Laramore 1992, Lotz 1992, Pantoja et al. 1999). In addition, Belak et al. (1999) reported that of 127 wild P. monodon collected off the coast of the Philippines, 50 individuals tested positive for IHHNV. They also noted that the prevalence of IHHNV infection in wild P. monodon is lower than in cultured stocks. A significant disadvantage of culturing wild-caught shrimp is the inability of the farmer to benefit from domestication and genetic improvement (Argue and Alcivar-Warren 2000). In the U.S. poultry industry, selective breeding has resulted in dramatic improvements in growth rate, feed conversion efficiency, and reproductive performance. For example, the chickens that we eat today grow twice as fast on half the amount of feed as the chickens of 50 years ago (Boyle 2000). This improvement is due, in large part, to the selective breeding practices of poultry breeders. Unfortunately, benefits accrued from the selective breeding of shrimp lag far behind those realized in other meat-producing industries. However, many penaeid shrimp possess characteristics amenable to selective breeding, including the ability to close the life cycle in captivity, a short generation time, and high fecundity (Moss et al. 1999). Clearly, it is in the shrimp farming industry’s best interest to use domesticated, genetically improved shrimp stocks that are free of specifically listed pathogens. This approach would improve production and profitability for the farmer, relieve pressures on wild-caught shrimp, and reduce problems associated with by-catch.

Use of Marine Protein in Feeds Although the poultry and swine industries are the largest consumers of fishmeal worldwide, fishmeal and fish oils often

Page 5 are dominant ingredients in aquaculture feeds because they serve as major sources of proteins and lipids, respectively, for carnivorous target species (Naylor et al. 2000). In semi-intensive and intensive aquaculture, two to five times more fish protein (in the form of fishmeal) is used to feed carnivorous finfish and marine shrimp than is supplied by the cultured product (Tacon 1996). This situation has led to charges that culturing aquatic carnivores depletes rather than augments fisheries resources (Goldberg and Triplett 1997, Naylor et al. 1998, Naylor et al. 2000). For shrimp, compound feeds typically contain 30% fishmeal and 3% fish oil. Using data from 1998, Tacon and Forster (2000) estimated that the conversion efficiency of pelagic fish to shrimp product is 2.18, so about two kg of pelagic fish are required to produce one kg of whole shrimp. This conversion efficiency is an order of magnitude greater than efficiencies estimated for several species of freshwater finfish, but not as high as for carnivorous marine finfish (Table 2). Table 2. Conversion efficiencies of pelagic fish to aquaculture species, based on a wet weight to wet weight ratio. Estimates assume a pelagic fish to fishmeal conversion ratio of 5:1.

Species

Conversion Efficiency

Salmon Trout Marine Shrimp Catfish Tilapia Carp

3.08 2.24 2.18 0.56 0.37 0.21

Data from Tacon and Forster (2000). While it is clear that the use of compound feeds to culture carnivorous species (including shrimp) currently results in the net consumption of marine protein, negative environmental impacts associated with the use of these feeds is debatable. Goldberg and Triplett (1997) have argued that the use of fishmeal in aquaculture “leads to a net loss of protein in a protein-short world”. However, fishmeal typically is made from small pelagic fishes, including menhaden, anchovy, pilchard, and capelin (Tacon 1996), and these fishes are not commonly consumed directly by humans (Suresh and Zendejas 2000). Goldberg and Triplett (1997) also have stated that there may be significant “ecological effects of massive harvests of small pelagic fishes”. While it is recognized that the removal of large quantities of pelagic fish would have a significant impact on the trophic ecology of marine ecosystems, the implication that aquaculture will play a major role in such a scenario is unwarranted. Chamberlain and Barlow (2000) have pointed out that, while aquaculture production has increased steadily over the past decade, annual worldwide fishmeal production

Page 6


has remained stable at 6-7 million metric tons, except for a decline during the 1998 El Niño period. They also argued that if fishmeal were to be completely eliminated from aquaculture feeds, pelagic fishes would continue to be harvested for use as fishmeal in terrestrial animal production. Irrespective of whether the negative environmental impacts resulting from the use of fishmeal in aquaculture are real or not, there is strong incentive for shrimp farmers and researchers to identify and develop alternative ingredients that can replace or supplement marine proteins. The cost of marine proteins is high relative to other ingredients (Tacon 2000), and their availability is unpredictable. Also, feeds containing marine proteins may serve as a vector for pathogens if they are not processed properly (Moss et al. 1998). Alternative ingredients include plants and their derivatives, rendered animal by-products, and microbial proteins (Suresh and Zendejas 2000, Tacon 2000, Smith et al. this volume). Recently, researchers at the Oceanic Institute demonstrated that shrimp fed a vegetable-based diet grew 1.06 g/wk and exhibited 81% survival over an 8-week trial (Argue et al. 2001). In a separate study, OI researchers determined that high-quality fishmeal can be completely replaced with either meat and bone meal or poultry by-product meal, without compromising shrimp performance (Tacon 2000). More research needs to be conducted to identify and evaluate alternative ingredients so that the demand for fishmeal and fish oils can be reduced.

EFFORTS TOWARD ENVIRONMENTALLY RESPONSIBLE SHRIMP CULTURE All economic activities have environmental impacts, including shrimp farming (Clay 1997). However, site-specific strategies can be adopted to mitigate many negative environmental consequences of shrimp aquaculture, thereby making the industry viable in the long-term. Over the past few years, recommendations and guidelines have appeared in the literature identifying environmentally responsible approaches to shrimp culture as they relate to site selection (Hajek and Boyd 1994, Boyd 1998), pond design and construction (Boyd 1998), feed and effluent management (Wang 1990, Hopkins et al. 1995b, Browdy et al. 1998, Calvo 1998, Lawrence et al. 1998), as well as seed source and quality (Wyban et al. 1993, Lotz et al. 1995, Pruder et al. 1995), and many of these approaches have been adopted by members of the global shrimp farming industry. Additionally, environmentally responsible codes of conduct have been drafted and formalized by various organizations with interests in fisheries and aquaculture, including the Food and Agriculture Organization (FAO) Fisheries Department Code of Conduct for Responsible Fisheries and the Global Aquaculture Alliance (GAA) Codes of Practice for Responsible Shrimp Farming (Boyd et al. this volume). Importantly, aquaculture associations from major shrimp farm-

ing countries of the world are adopting environmentally responsible codes of conduct, including Thailand (Fegan 1999) and Ecuador (Chamberlain 1999). These codes identify guidelines and best management practices (BMPs) for each component of the value chain in an attempt to realize the goal of establishing an environmentally, socially, and economically responsible shrimp farming industry. In the following section, we do not provide a comprehensive overview of all published recommendations and guidelines for environmentally responsible shrimp culture. Rather, we provide a discussion of two critical approaches to shrimp culture that mitigate many of the negative environmental impacts of shrimp farming while simultaneously increasing production and profitability for the farmer. These approaches should be considered in all BMPs drafted for the shrimp aquaculture industry.

Reduced Water Exchange In the recent past, conventional wisdom dictated that there is a positive relationship between water exchange rate and shrimp production, and evidence exists in the literature to support this contention. In Asia, production of Penaeus monodon improved from 10 mt/ha/yr to 25 mt/ha/yr when the maximum water exchange rate increased from 20% to 100% per day (Hirasawa 1985). However, due to concerns about nutrient and biological pollution of the natural environment, and self-pollution and disease introduction into the shrimp culture environment, high-volume water exchange is no longer considered an appropriate pond management strategy in many shrimp farming regions of the world. Importantly, over the past decade, research results indicate that high rates of water exchange are not necessary to maintain high shrimp biomass under intensive culture conditions. For example, Browdy et al. (1993) examined the effects of different water exchange rates (10, 50, and 100% exchange/day) on growth and survival of Litopenaeus vannamei stocked in outdoor tanks at two densities (60 and 100 shrimp/m2). Exchange rate and density had little impact on water quality, survival, and growth in all treatments except the high density (100 shrimp/m2), low water exchange (10% exchange/day) treatment, which experienced low dissolved oxygen (DO) concentrations resulting in high shrimp mortality. Results from this study suggest that reduced water exchange rates do not compromise shrimp growth or survival under intensive culture conditions, if acceptable DO concentrations are maintained. In an effort to further reduce water exchange rates, Hopkins et al. (1991) examined the effects of two water exchange rates (4 and 14% exchange/day) on the production of L. vannamei stocked in outdoor ponds at a density of 76 shrimp/m2. Results from this study corroborate those of Browdy et al. (1993) and indicate that there was no significant effect of water exchange rate on shrimp growth or survival at intensive stock-

Moss ing densities. In fact, production in the 4% exchange/day treatment (7,565 kg/ha/crop) was slightly higher than in the 14% exchange/day treatment (7,462 kg/ha/crop). In a subsequent study to determine the impacts of zero water exchange, Hopkins et al. (1993) examined the effects of normal (25% exchange/day), reduced (2.5% exchange/day) and zero water exchange on the production of L. setiferus stocked in 0.1 ha ponds at a density of 44 shrimp/m2. In addition, zero exchange ponds were stocked at 22 and 66 shrimp/m2 to examine density effects under zero-exchange conditions. Production in the normal and reduced water exchange treatments was similar at 5,718 kg/ha/crop and 6,375 kg/ha/crop, respectively, whereas production in the low density (22 shrimp/m2), zero exchange treatment was 3,219 kg/ha/crop. However, the combination of higher stocking density (44 and 66 shrimp/m2) and zero water exchange resulted in high shrimp mortality that was attributed, in part, to gill fouling. Results from this study indicate that water exchange in intensive shrimp culture can be eliminated without compromising shrimp growth or survival, if stocking densities less than 44 shrimp/m2 are used. Importantly, by reducing or eliminating water exchange, the amount of nutrients, solids, and biological oxygen demand (BOD) discharged into receiving waters during the production process (i.e. daily water exchange and drain harvest) can be reduced significantly due to the metabolic activities of in situ microorganisms. The term “microorganism” generally refers to bacteria, microalgae, fungi, yeast, and protozoans (Maeda 1999), and these organisms play a critical role in maintaining acceptable water quality by providing oxygen, converting toxic metabolites to less toxic forms, and oxidizing organic matter. In addition to their role in affecting water quality, microorganisms have a profound and direct impact on shrimp growth and survival. For example, pond microbes and associated detritus can enhance shrimp growth significantly by serving as a nutritional source (Leber and Pruder 1988, Moss et al. 1992, Moss 1995, Moss and Pruder 1995). Also, pond water (containing microbes and detritus) can affect the abundance and species composition of shrimp gut flora (Moss et al. 2000), stimulate digestive enzyme activity in shrimp digestive glands (Moss et al. in press), and enhance the non-specific immune response in shrimp by stimulating serum agglutination (Primavera et al. 2000a). The assimilative capacity of microbes to process and recycle organic matter and remove toxic metabolites from the culture environment is determined largely by the quantity and quality of exogenous feed supplied to the system (Brune and Drapcho 1991). It is likely that the upper limit of shrimp production in zero exchange systems also is significantly affected by exogenous feed input, especially feed protein content

Page 7 (Hopkins et al. 1995b, Browdy et al. 1998). This assertion is supported by results from Hopkins et al. (1995a), who examined the effects of two diets (20 and 40% protein) on growth and survival of L. vannamei stocked in zero exchange ponds at two densities (39 and 78 shrimp/m2). Production in the low protein (20%), high density (78 shrimp/m2) treatment (8,163 kg/ha/crop) was greater than in the other three treatments, and was 154% greater than the maximum production reported in Hopkins et al. (1993) under zero exchange conditions. Importantly, there were no obvious differences in harvest weight between shrimp fed the two protein diets. It is likely that the endogenous production of natural productivity in the 20%-protein treatments provided shrimp with sufficient nutrients to compensate for nutritional deficiencies associated with this lower protein feed (see Otoshi et al. 2000). Researchers continue to develop and evaluate shrimp diets that are appropriate for zero exchange systems, and “environmentally friendly” feeds are being developed to minimize pollution during daily water exchange or drain harvest (Lawrence et al. 1998, Tacon et al. 2000, Lawrence et al. this volume). Also, improved feed management strategies (e.g. use of feed trays) offer additional opportunities to maximize shrimp performance while simultaneously minimizing water quality problems associated with exogenous feed inputs (Jory 1995). Although research results from zero exchange studies appear promising, few commercial shrimp farms are able to achieve high production under zero exchange conditions. One notable exception is Belize Aquaculture, Limited, which has taken the lead in developing a zero exchange system using low protein feeds and heavy aeration (McIntosh 1999). Similar to the zero exchange systems described above, the Belize system takes advantage of in situ microbes for biofiltration and as a source of supplemental food for the shrimp. However, unlike these other systems, there is a predictable shift in microbial community structure over the course of the growout period, where mixed photoautotrophic/heterotrophic microbial communities are replaced by bacterial “floc” which maintains system stability and enables the pond to assimilate large amounts of organic inputs. During the initial 23 months of operation, mean production at Belize Aquaculture was 11,231 kg/ha/crop (McIntosh 1999), and this is 38% greater than the maximum production reported in Hopkins et al. (1995a) under zero exchange conditions. Importantly, the Belize system releases very little nutrients or organic matter into the surrounding environment because pond effluent is directed to a settling basin prior to re-use or eventual discharge. The success of Belize Aquaculture represents a significant milestone for the shrimp aquaculture industry and their approach to shrimp culture reflects an important paradigm shift from traditional shrimp farming methods. However, the production systems at Belize Aquaculture are not biosecure and


Page 8

the potential for a significant disease outbreak exists. Biosecure shrimp production systems represent an emerging technology that provides a high degree of pathogen exclusion with minimal water exchange (Moss et al. 1998, Lotz and Lightner 2000). In an effort to develop biosecure technologies for the U.S. shrimp farming industry, the Oceanic Institute is evaluating a prototype biosecure system for the intensive production of L. vannamei. The system consists of a covered, 58-m2 raceway that allows for the continuous flow of water in a circular pattern around a central baffle (Moss 1999). The flow of water produces a scouring velocity to keep solids in suspension. An aspirator-type aerator is used to provide both aeration and water movement. For water filtration, the system relies on a 25ft3 propeller-washed bead filter (PBF) for solids removal and biofiltration, and the system requirement for water replacement is less than 0.5%/day. Importantly, the PBF allows a sufficient amount of microalgae and other microbes to pass through it so that a “green water” environment can be maintained. In this type of environment, shrimp can take advantage of the growth enhancing effect of pond water described earlier. In a recent trial, juvenile shrimp were stocked at a density of 200/m2 and were grown to a harvest weight of 23.3 g in 16 weeks (Moss et al. 2001b). During this trial, shrimp grew by 21.1 g, growth rate was 1.32 g/wk, survival was 85%, and production was 4.0 kg/m2 (40,000 kg/ha/crop). This level of production is 256% greater than the mean production reported at Belize Aquaculture (McIntosh 1999), and 106% greater than the maximum production reported in McIntosh and Carpenter (1999) under commercial, zero exchange conditions. Importantly, the amount of water used to produce one kg of whole shrimp was about 370 liters, and this is two to three orders of magnitude less than what is commonly used by the existing shrimp farming industry. According to Hopkins and Villalón (1992), the amount of water used by shrimp farmers to produce one kg of whole shrimp ranges from 39,000 to 199,000 liters, with a mean value of 86,000 liters. Information used to determine these values came largely from semiintensive shrimp farms and these values include exchange water

plus one pond volume for drain harvest. In more intensive systems with minimal water exchange, water use per kg of shrimp produced can be reduced significantly (Table 3). The importance of reduced or zero water exchange cannot be understated. In addition to reducing nutrient and biological pollution of the natural environment, this approach minimizes self-pollution of the shrimp culture environment and reduces the opportunity for pathogen introduction. Results from research designed to investigate reduced or zero water exchange indicate that high shrimp biomass can be supported under these conditions, if appropriate feeds and aeration are provided. Importantly, commercial shrimp farmers around the world are adopting this strategy and results are encouraging (McIntosh 1999, Fegan 2000, Hamper 2000).

Specific Pathogen Free (SPF), Selectively Bred Shrimp Most shrimp cultured worldwide are collected from the wild or are offspring from wild-caught broodstock. As we mentioned earlier, this practice is risky for shrimp farmers because the disease status of wild shrimp often is unknown and wild shrimp may be carriers of pathogens. Importantly, the global shipment of wild shrimp (nauplii, postlarvae, and broodstock) occurs routinely because there is a need to provide seed to a dependent shrimp farming industry. Unfortunately, this practice has resulted in the transfer and introduction of shrimp pathogens from one geographical region to another with devastating economic consequences (Lightner et al. 1992). In an effort to reduce the accidental introduction of nonnative organisms (including pathogens), guidelines have been established to assist governments and the private sector in importing non-native aquatic organisms for fishery or aquaculture use in a responsible manner. The International Council for the Exploration of the Sea (ICES) has adopted a “Code of Practice to Reduce the Risks of Adverse Effects Arising from

Table 3. Amount of water used to produce one kilogram of whole shrimp. Data are from commercial shrimp farms and research institutions that use reduced or zero water exchange under intensive culture conditions.

Shrimp Species Litopenaeus setiferus L. setiferus L. setiferus L. vannamei L. vannamei L. vannamei L. vannamei

Water Exchange/Day (%) 25.0 2.5 0 0 < 10.0 < 0.5 < 0.5

Stocking Density (#/m2)

Water use/kg shrimp (liters/kg)

40 40 20 63 – 121 35 100 200

64,000 9,000 6,000 2,000 1,500 483 370

Reference

Hopkins et al. (1993) Hopkins et al. (1993) Hopkins et al. (1993) Fast and Menasveta (2000) Hamper (2000) Moss et al. (2001b) Moss et al. (2001b)

Moss

Page 9

the Introduction of Non-indigenous Marine Species” (Sindermann 1990), and modifications of these ICES guidelines have been used to develop specific pathogen free (SPF) stocks of penaeid shrimp for the U.S. Marine Shrimp Farming Program (Wyban et al. 1993, Pruder 1994, Pruder et al. 1995, Lotz et al. 1995). The guidelines stipulate that only diseasecausing organisms that can be reliably diagnosed and physically excluded from a facility can be considered in an SPF program. Currently, the list of excludable pathogens for shrimp includes nine viruses or virus groups, certain classes of parasitic protozoans, and metazoan parasites (Table 4). It is likely that this list will be revised and expanded as new pathogens are described and more advanced disease diagnostic tools become available. To develop an SPF stock, shrimp are collected from the wild and transferred to a primary quarantine facility where they are analyzed for specifically listed pathogens using appropriate disease diagnostic tools (Figure 1). If shrimp test positive for any of the listed pathogens, they are destroyed in the primary quarantine facility. If shrimp test negative for specifically listed pathogens after several successive screenings, they are transferred to a secondary quarantine facility where they are matured and spawned to produce an F1 generation of captive shrimp. Because some viruses can be transmitted from parent to offspring (vertical transmission), representative shrimp from the F1 generation are analyzed for specifically listed pathogens. If shrimp from the F1 generation test negative for specifically listed pathogens after several successive screenings, they are transferred out of the secondary quar-

Wild Shrimp Stocks

Primary Quarantine 2-5 months

YES

Candidate “SPF” stocks

Secondary Quarantine & Maturation 5-12 months

NO

Candidate “SPF” stocks

NO DESTROYED

YES

Nucleus Breeding Center

Figure 1. Procedures used to stock the SPF-NBC at the Oceanic Institute (modified from Carr et al. 1994).

Table 4. A working list of “specific” and excludable pathogens for penaeid shrimp.1

Pathogen Type Viruses

Protozoa

Metazoan Parasites

Pathogen/Pathogen Group WSSV - “whitespotviridae” the white spot syndrome baculo-like viruses YHV, GAV, LOV – systemic cytoplasmic rhabdo-like viruses TSV – a picornavirus BPV- an occluded enteric baculovirus MBR – an occluded enteric baculovirus BMN - a nonoccluded enteric baculovirus IHHNV – a systemic parvovirus SMV – an enteric parvo-like virus HPV – enteric parvovirus Microsporidians Haplosporidians Gregarines Larval nematodes Larval trematodes Larval cestodes

Pathogen Category2 C-1 C-1 C-1 C-2 C-2 C-2 C-1 C-1 C-2 C-2 C-2 C-3 C-3 C-3 C-3

1

Modified from D.V. Lightner, U.S. Marine Shrimp Farming Program FY00 Progress Report.

2

Pathogen category (modified from Lotz et al. 1995) with C-1 pathogens defined as excludable pathogens that can potentially cause catastrophic losses in one or more penaeid species; C-2 pathogens are serious, potentially excludable; and C-3 pathogens have minimal effects, but may be excluded from NBCs, multiplication facilities, and some types of farms.

Page 10


antine facility and can be included as part of the germplasm in a nucleus breeding center (NBC). Shrimp that are maintained in a well-established NBC (i.e. where there is a history of negative disease status documented through a surveillance program) may be designated as SPF (Lotz 1997). However, once shrimp leave an SPF-NBC, they no longer are referred to as SPF even though they may be free of specifically listed pathogens. The new designation is High Health (HH) shrimp, and this indicates that these shrimp are at greater risk of pathogen exposure and infection. Early growout trials using HH L. vannamei indicated that these stocks out-performed non-HH stocks when evaluated at commercial shrimp farms in the United States (Wyban et al. 1993). In Texas, Jaenike et al. (1992) reported that HH shrimp obtained from the U.S. Marine Shrimp Farming Program produced a greater yield, higher survival, a more uniform size distribution, and a lower feed conversion ratio than non-HH shrimp. In Hawaii, Carpenter and Brock (1992) reported that HH shrimp produced a greater yield and higher survival than non-HH shrimp when cultured under semi-intensive and intensive culture conditions. Importantly, the HH crop yielded a 62.5% higher return than non-HH crops, and similar improvements were reported in South Carolina (Wyban et al. 1993). Despite these encouraging results, SPF and HH shrimp were not considered a panacea to the disease problems plaguing the shrimp farming industry (Pruder 1994). In 1993, HH shrimp were cultured with wild-caught seed at a commercial shrimp farm near Rio Guayas in Ecuador. HH shrimp exhibited poor survival (7-43%) compared to wild seed (36-42%), and heavy mortalities were attributed to TSV. Clearly, SPF or HH shrimp cultured in environments experiencing disease problems may not perform well. It is possible that improved performance can occur if stocks are genetically modified through such mechanisms as natural or artificial selection. For example, some shrimp farmers breed survivors of specific disease outbreaks with the expectation that the offspring will possess characteristics capable of conferring protection against the pathogen of concern. These shrimp are referred to as specific pathogen resistant (SPR) stocks (Lightner 1995), and such stocks have been developed in Polynesia where IHHNV-resistant L. stylirostris have been produced. However, SPR stocks may be asymptomatic carriers of viral pathogens, so there is strong incentive to develop SPF/SPR strains of shrimp for the global shrimp farming industry (see below). The importance of using SPF or HH stocks can not be understated, as they offer clear advantages over diseased stocks or stocks with an undetermined disease status. For example, the commercial hatchery at Harlingen Shrimp Farms in Texas experienced only one viral outbreak during its nine-year history of using broodstock derived from SPF stocks (Jaenike and Page 1999). The use of these shrimp, coupled with other biosecurity protocols, has al-

lowed Harlingen Shrimp Farms to produce high-health seed for domestic and international markets. A major advantage in using captive shrimp is the opportunity to benefit from domestication and selective breeding. The process of domestication allows for the selection of shrimp that are better adapted to the artificial conditions within which they are cultured. However, it is important to note that domestication does not always improve desired aquacultural traits (Doyle 1983). The application of selective breeding techniques offers tremendous opportunity for increased production and profitability for the shrimp farmer by improving traits of economic interest, such as growth and disease resistance (Gjedrem and Fimland 1995). Surprisingly, domestication and selective breeding programs for shrimp are not common despite encouraging research results (see below). A recent issue of the Global Aquaculture Advocate (1999, volume 2, issue 6) highlighted several domestication/breeding programs, including those in Thailand, New Caledonia, Tahiti, Hawaii, Belize, Venezuela, Australia, and Colombia. However, most of these programs are associated with research institutions or are at the early stages of commercial development. As the economic benefits of genetic improvement become more compelling, it is likely that the global shrimp farming industry will invest in selective breeding programs that rely on SPF stocks. The Oceanic Institute (OI) has played a significant role in establishing some of the fundamental principles of a selective breeding program for shrimp. In 1989, the U.S. Marine Shrimp Farming Program, with funding from the U.S. Department of Agriculture, developed the first population of SPF L. vannamei at OI for distribution to U.S. shrimp farmers (Wyban et al. 1993). OI has honored the spirit and intent of the ICES guidelines (see above) in establishing founder stocks for the breeding program. To date, OI has obtained SPF populations of L. vannamei from six different regions of its natural range, including Mexico, Ecuador, and Panama. New populations will be added to the program in the future to increase the genetic diversity of the breeding stocks. OI maintains an SPF-NBC at its Makapu’u, Hawaii facility where SPF L. vannamei broodstock are used to produce 80 full-sib families twice a year. Broodstock are selected based on a breeding value assigned to families (family selection). A female from a specific family is mated with a specific male by artificial insemination. In order to minimize inbreeding, there are no sibling or cousin matings. Following insemination, the female is placed in an individual spawning tank where fertilized eggs are liberated. Viable nauplii are then transferred to individual containers in an indoor hatchery facility where they are reared to postlarvae. Postlarval shrimp are transferred outdoors to individual nursery tanks where they are grown to about one gram, at which time they are injected with an inter-

Moss

Page 11

nal, elastomer tag to identify the family from which they came (Godin et al. 1996). Once the shrimp are tagged, they can be cultured in a common environment and inferences can be made about family performance. From 1995-1998, shrimp were selected based on an index where equal emphasis was placed on growth and resistance to TSV (Table 5). Since 1998, two separate lines have been established. One line is selected 70% for TSV resistance and 30% for growth in flow-through shrimp ponds, whereas the other line is selected 100% for growth in the recirculating raceway referred to in the previous section. Shrimp have been evaluated for growth at various research and commercial facilities in Hawaii and on the U.S. mainland, whereas disease-challenge tests have been conducted at research labs in Arizona, Mississippi, and South Carolina. Prior to the inception of OI’s breeding program, there was a paucity of information about quantitative genetics of penaeid shrimp. One of our continuing goals is to generate information about genetic parameters relevant to a breeding program, including estimates of heritability (h2), phenotypic and genetic variation, phenotypic and genetic correlation, and genotypeenvironment interaction. To date, over 600 families have been evaluated for growth and TSV resistance, and information about h2, phenotypic variation, and genetic correlation of these traits is summarized below. Heritability describes the percentage of phenotypic variance that is inherited in a predictable manner and is used to determine the potential response to selection. Theoretically, h2 ranges from 0 to 1, although calculated values can exceed

this range. Traits with h2 estimates approaching 1 are highly heritable, whereas traits with h2 estimates approaching 0 are not heritable. Data generated from OI since 1995 indicate that the mean h2 estimate for weight gain is 0.40 + 0.06 (h2 half-sib + SE, Argue et al. 1999a). This estimate is considered moderate to high, and significant improvements in this trait should be made through selection. In contrast, the mean h2 estimate for TSV resistance is only 0.09 + 0.03 (h2 half-sib + SE, Argue et al. 1999b). Estimates of h2 typically are low for fitness traits, such as disease resistance, and phenotypes with h2 estimates less than 0.15 are often difficult to improve by selection (Tave 1993). In addition to generating h2 estimates for growth and TSV resistance, OI researchers have calculated h2 estimates for sex ratio, percent tail, NH3 tolerance, and growth on vegetable-based diets (Table 6). Although h2 estimates for TSV resistance are low, there is high between-family variation in response to a TSV challenge. For example, family survival after 14 days post-exposure to TSV per os ranged from 0% to 88% in a generation of shrimp produced in 1997 (Figure 2). This degree of phenotypic variation is common in every generation of selectively bred shrimp produced at OI that has been subjected to a TSV-challenge test. In addition to h2 estimates and information about phenotypic variation, the relationship between TSV survival and harvest weight has been established. Obtaining correlations among commercially important traits is especially important in breeding programs that rely on multiple trait selection.

Table 5. Summary of the shrimp breeding program at the Oceanic Institute from 1995 – 2000.

Year

Batch #

1995 1995 1995 1996 1996 1997 1997 1998

1 2 3 4 5 6 7 8

1998 1999 1999 2000

9 10 11 12

1 = Sinaloa, México 2 = Ecuador 3 = México x Ecuador hybrids

Selection Criteria 50% Growth / 50% TSV 50% Growth / 50% TSV 50% Growth / 50% TSV 50% Growth / 50% TSV 50% Growth / 50% TSV 50% Growth / 50% TSV 50% Growth / 50% TSV 100% TSV 100% Growth 50% Growth / 50% TSV 100% Growth 30% Growth / 70% TSV 100% Growth 30% Growth / 70% TSV 4 = Oaxaca, México 5 = Chiapas, México

Populations 1, 2, 3 1, 2, 3 1, 2, 3 1, 2, 3 1, 2, 3, & 4 1, 2, 3, 4, & 5 1, 2, 3, 4, 5, & 6 1, 2, 3, 4, & 5 1, 2, 3, 4, & 5 1, 2, 3, 4, & 5 1, 2, 3, 4, 5, & 6 1, 2, 3, 4, 5, & 6 1, 2, 3, 4, 5, 6, & 7 1, 2, 3, 4, 5, 6, & 7 6 = Esmeraldas, Ecuador 7 = Panamá hybrids


Page 12

Table 6. Half-sib heritability estimates (h2half-sib) of commercially important traits for Litopenaeus vannamei.

h2half-sib (SE)

Trait Sex Ratio Percent Tail NH3 Tolerance Growth on Vegetable Diets

-0.002 0.15 0.16 0.40

± 0.012 ± 0.12 ± 0.10 ± 0.30

10 0

% SURVIVAL

80 60 40 20 0 FAM ILY

Figure 2. Family survival at 14 days post-exposure to TSV per os. Mean % survival for all 80 families was 38%. Results from our research indicate that there is a significant negative correlation between mean family survival in the TSVchallenge test and shrimp harvest weight from a growout pond at OI (r = -0.45, p