Impact of a once-through cooling system on the yellow perch stock in the western basin of Lake Erie.Ecological .... Fishing News Books, Oxford, United Kingdom.
Estuaries
Vol. 18, No. 1A, p. 25-42
March 1995
Environmental Impacts of Shrimp Farming with Special Reference to the Situation in the Continental United States j . STEPIIEN I|OPKINS 1 PAUL A. SANDIFER
Waddell Ma~cullure Center P.O. Box 809 Bluffion, South Carolina 29910 M. R[ClIAm) DEVOE South Carolina Sea Grant Consortium 287 Meeting Street Charleston, South Carolina 29401 A. FREI)ERICK l IOLIA.X'D CRAIG L. BROWDY ALV:N D. STOKES Waddell Mariculture Center P.O. Box 809 Bluff ton, South Carolina 29910 ABSTRACT: Shrimp culture technology has resulted in development of a major shrimp farming industry worldwide. Without the shrimp farming industry, increasing demands for shrimp by consumers could not be met, resulting in increased pressure on wild shrimp resources. Unfortunately, there are realized and potential adverse environmental effects on estuarine ecosystems as a result of shrimp farming. The effects can be categorized as wetland destruction for construction of shrimp farms, hypernutrificatJon of estuarine ecosystems by shrimp pond effluent, "biological pollution" of native shrimp stocks through escapement of aquaculture stocks, water use and entrainment of estuarine biota, and impacts of shrimp farm chemicals on estuaxine systems. While the shrimp farming industry in the United States is small, the United States is effectively addressing all the realized and potential environmental impacts through regulation and research at the federal and state levels. Areas of regulation and research include stringent prohibitions on wetland destruction, regulation of effluents and support of research to eliminate a n d / o r reduce effluents, escapement prevention technology and development of high-health stocks, minimizing entrainment of estuarine biota through water conservation and screening technology, and regulation of chenfical use in the shrimp farming industry and support of research on shrimp pathology and environmentally safe disease control. Work is still in progress and not all problems have been resolved to the complete satisfaction of shrimp farmers and estuarine conservationists. However, the situation in the United States should serve as a model of how to encourage sustainable economic development through commercial shrimp farming while abating adverse environmental impacts on estuarine systems. To further improve the situation, the development and adoption o f "best management practices" for shrimp aquaculture are recommended.
An I n t r o d u c t i o n to Shrimp F a r m i n g Technology
m e n t to e n h a n c e shrimp p r o d u c t i o n . Systems that o p e r a t e with limited control, limited m a n a g e m e n t inputs, a n d low p r o d u c t i o n goals are t e r m e d extensive, while m o r e "intensive" systems e x e r t greater control for higher p r o d u c t i o n . Extensive systems are often built in wetlands where land costs are low a n d p o n d s are tilled a n d drained with tidal action. Extensive s h r i m p farming takes advantage of wild shrimp p o s d a r v a e r e c r u i t e d into the system, possibly with s o m e enh a n c e m e n t , and relies largely on naturally-occur-
A wide array o f systems has b e e n used to farm p e n a e i d shrimp. At o n e e n d o f tile s p e c t r u m are i n d o o r tanks where m o s t aspects o f e n v i r o n m e n t and n u t r i e n t are controlled. At the o t h e r e x t r e m e are large sections of estuary that have b e e n partitioned off t h r o u g h natural processes or m a n ' s intervention a n d that receive r u d i m e n t a r y manage-
Corresponding author. (c) 1995 Estuarine Research Federation
25
0160-8347/95/01A0025-18501.50/0
26
J.S. Hopkins et al.
ring forage items for shrimp nutrition. T h e p o n d biota is tairly diverse, and may nearly mimic that of the estuary, h l c l u d e d in this biota are limited n u m b e r s o f tish, which prey on the shrimp crop. Little feed or fertilizer is used and the water quality in the p o n d may be similar to that o f the adjacent estuary. Crop yields are only a few h u n d r e d kg ha -l crop ~, but operating costs are also low (Whetstone et ai. 1988; Fast 1992). Semi-intensive systems employ ponds built at a high e n o u g h elevation to be drainable. They rely on stocking known numbers of postlarvae, which are collected in the estuary or reared in a hatchery. Semi-intensive systems may infringe on the u p p e r portions of wetlands, but water is typically p u m p e d into the p o n d and drained by gravity back to the estuary. As water is p u m p e d into the pond, it is screened to exclude p r e d a t o r y fish. Inexpensive shrimp feed is provided, but much of the shrimp nutrition is derived fi-om naturally-occurring ti)rage items. Fertilizers are used to increase primary productivity and e x p a n d the forage base. T h e semiintensive p o n d biota is less diverse than that o f the estuary or extensive ponds, but certain zooplanktonic and benthic species are abundant. Because of fertilizer and feed inputs, semi-intensive p o n d water is high in organic matter and nutrients and has suspended solids and BOD concentrations higher than the adjacent estuary. T h e nutrient inputs may be high e n o u g h to create occasional problems with dissolved oxygen depletion in the pond, and e m e r g e n c y aeration e q u i p m e n t may be employed as needed. Crop yields are less than 2 metric tons ha ' crop -1 (Clifford 1985). Intensive systems use smaller ponds built on high ground, away ti'om wetlands, so the p o n d bottoms can be dried and manipulated between crops. Large n u m b e r s of hatchery postlarvae are stocked. I,arge zooplankters and benthic species are rapidly grazed by the dense population of shrimp (Hopkins et al. 1988). Much o f the shrimp nutrition is supplied by expensive, nutritionally complete rations. These high-protein shrimp feeds act as organic fertilizers, so intensive ponds are very rich in nutrients. Dissolved inorganic nutrients p r o m o t e dense blooms of phytoplankton. T h e p o n d is eut r o p h i c a n d dissolved o x y g e n c o n c e n t r a t i o n s would often decline to lethal levels if not supplem e n t e d by aeration equipment. Once oxygen dem a n d is satisfied, the n e x t limiting factor is toxic metabolites such as ammonia. Crop yields for intensive systems may be as high as 20 mt ha -~ crop -~ (Sandifer et al. 1991). Commercial aquaculture of shrimp occurs almost exclusively in o u t d o o r ponds. T h e r e have been attempts to p r o d u c e shrimp in i n d o o r systems, some of which had sophisticated filtration
and almost complete water reuse. Some have been biologically successfid and p r o d u c e d shrimp at astounding densities; however, n o n e has been economically successful due to the high operating costs. As t e c h n o l o g y has b e e n r e l i n e d , i n d u s t r i a l shrimp farming has b e c o m e a reality and competition has increased. Competition has the effect o f selecting for the p r o d u c t i o n technology that is most profitable within a given set of geographical and sociological conditions. Thus, most shrimp farlns in Taiwan have similar intensive p r o d u c t i o n ponds (Hopkins 1986; I.iao 1992). In Ecuador, there are vast areas of large semi-intensive ponds (Aiken 1990). In Viet Nam, extensive shrimp ponds are often built intertidally wherever topography permits (Vu 1992). The United States Position in a Global Perspective
O f the 2.4 million metric tons of shrimp annually placed into the world market, 560,000 mt or 25% is cultivated on farms rather than captured in the ocean (Rosenberry 1990). While commercial harvest of wild shrimp stocks has stabilized or is declining slightly due to overfishing and habitat loss, shrimp farming continues to expand. Most shrimp consumers are not aware that shrimp farming p r o d u c t i o n technology has largely stabilized the price o f shrimp over the past decade. Without shrimp farms, increased d e m a n d would result in dramatic increases in prices for the finite wild stocks, thus increasing fishing pressure. If consumer d e m a n d for shrimp continues to rise, virtually all of the additional p r o d u c t will c o m e from tiarms. At current rates o f increase, shrimp farms could be p r o d u c i n g 50% o f a 3 million mt market by the year 2000 (Rosenberry 1990). T h e United States is a major market outlet for shrimp from a r o u n d the world. While about 365,000 mt o f shrimp are consumed annually in the United States, domestic shrimp fisheries annually catch only 92,500 mt (25%). Most of the remaining 75% is i m p o r t e d farm-raised shrimp (Hopkins 1992). Although the United States is a major c o n s u m e r of shrimp and has made substantial contributions to d e v e l o p m e n t o f shrimp farming technology, the domestic shrimp farming industry is relatively small. Projected 1992 total p r o d u c t i o n of farmraised shrimp in the continental United States is only 1,800 mt (personal connnunication G. Treece, Texas A&M Sea Grant College Program; A. Stokes, Waddell Mariculture Center), 1% o f the deficit not supplied by domestic fisheries. T h e federal government, through the United States D e p a r t m e n t o f Agricuhure, has identified the national trade deficit in shrimp as an item o f c o n c e r n and is currently
Environmental Impacts of Shrimp Farming
funding programs that seek to decrease that deficit t h r o u g h expansion of the domestic shrimp farming industry. T h e degree o f expansion is depend e n t on a variety of interrelated factors. Profitability and the availability of investment capital are of major importance, but regulatory constraints and unbridled expansion of shrimp I~trming in o t h e r counties are key factors as well. Technological improvements to p r o d u c t i o n processes will undoubtedly reduce costs and be reflected in the price consumers pay. However, rapid dissemination of new information and techniques to o t h e r parts of the world act to buffer the technological advantage on expansion o f the domestic shrimp farming industry.
Environmental Impacts of Shrimp Farming In the early phases ot" its development, shrimp farming and aquaculture in general were t h o u g h t to be c o m p l e t e l y " c l e a n " i n d u s t r i e s (Weston 1991). Slowly, this perception is changing as overd e v e l o p m e n t o f shrimp farming industries in localized areas a r o u n d the world begins to create noticeable changes in the natural environment. Most of the available data on the environmental impacts o t aquaculture have c o m e from studies on the netpen c u h u r e of salmonids. Only in the past few years have some o f the environmental consequences of shrimp farming been seriously considered. Realized and potential environmental impacts can be categorized as destruction of wetlands for shrimp p o n d construction, hypernutrification of estuaries by shrimp p o n d effluents, ecological disruption of indigenous shrimp stocks through postlarvae and broodstock acquisition and expansion o f the range of non-indigenous shrimp species or shrimp diseases through livestock transfers, water use and e n t r a i n m e n t o f estuarine biota, and effects of shrimp farming chemicals on adjacent estuarine ecosystems. WETIAND DESTRU(YIION FOR SHRIMP POND CONSTRUCTION
Globally, the most i m p o r t a n t adverse environmental effect o f shrimp farming is the destruction of wetland areas, primarily mangrove swamps, tor s h r i m p p o n d c o n s t r u c t i o n ( L e e a n d Wickins 1992). Mangroves are o f vital i m p o r t a n c e to their estuarine ecosystems ( O d u m and Heald 1972). An estimated 19% of the 170,000 ha of mangrove swamps in Ecuador have b e e n destroyed through shrimp farm construction over two decades (Aiken 1990). In one area of Bangladesh, 34% o f the mangrove forest was converted to fish and shrimp ponds b.etween 1974 and 1986 (Shahid and Pramanik 1986). In the Philippines, conversion of mangroves to brackish water farms has been in
27
progress tot centuries, and an estimated 206,500 ha, representing about 45% of the original mangroves, had been t n r n e d into milkfish and shrimp ponds by 1984 (Pillay 1992). However, these figures do not differentiate between supratidai and the m o r e i m p o r t a n t intertidal mangrove-dominated land. T h e significance o f the two types in terms of their contribution to estuarine ecosystems are dramatically different. In Indonesia, p o n d develo p m e n t is prohibited in a 100 m wide band o f mangroves along the coast. Indonesian managers are willing to c o m m i t 10-15% of their mangrove area to aquaculture (Chamberlain 1991). T h e r e are panels to govern mangrove m a n a g e m e n t in Malaysia, the Philippines, and Thailand as well (McVey 1988). In m o r e temperate zones, cordgrass and spikerush marshes can also be affected. Whetstone et al. (1988) r e p o r t that shrimp p r o d u c t i o n is the most feasible type o f brackish water aquaculture in almost 29,000 ha of coastal marsh that were i m p o u n d e d prior to inception of coastal zone m a n a g e m e n t programs in South Carolina. T h e impetus for wetland destruction is often low land cost and low operating costs for p u m p i n g water. Shrimp farming is o n e of the few industries that can profitably utilize wetlands, so there is little competition for space from o t h e r interests. In some situations, construction costs are'. also low. N a t u r a l s h o r e l i n e c o n t o u r s may m i n i m i z e the earth-moving requirements to i m p o u n d large areas of wetland. Extensive p r o d u c t i o n in tidal ponds does not reqnire capital for pumping, feeding, or, in some cases, stocking. However, p o n d construction at higher elevation provides better water control, avoids problems with acid-sulfate soils, typically involves less clearing o f vegetation, and may facilitate.- use of heavy equipment, which actually lowers construction costs (McVey 1988; P o e r n o m o
1990). T h e r e is a definite irony in destroying wetlands for shrimp p o n d construction. As productive wetlands are removed fl'om the estuarine system, wild shrimp populations decline ( T u r n e r 1977) and the ability to recruit a n d / o r catch juvenile shrimp for stocking p o n d s may be reduced. Destruction o f mangrove nursery habitat for p o n d construction in Ecuador was perceived as a factor in the decline in n u m b e r s o f postlarvae that could be captured for p o n d stocking (Twilley 1989). I,ocal fisheries for shrimp and o t h e r species are impacted. Furthermore, converted wetlands d o not generally make good shrimp ponds. They are either impossible to drain and thus c a n n o t be effectively managed, or are difficult to.drain and u n d e r g o an acid-sulfate reaction. Acid-sulfate soils are f o r m e d where organic matter is c o n c e n t r a t e d in waterlogged conditions, such as wetland areas. In the anaerobic
28
J.S. Hopkins et al.
conditions, sulfate-reducing bacteria d e c o m p o s e organic matter and release sulfide. When these soils are exposed to air during p o n d construction and p o n d m a n a g e m e n t , aerobic bacterial oxidation produces sulturic acid and an orange-red ferric hydroxide precipitates on the surface. T h e sulfuric acid decreases soil p H and ferric hydroxide has been f o u n d to clog shrimp gills (Nash et al. 1988; Lee and Wickins 1992). Coastal d e v e l o p m e n t is r a m p a n t in many, if not most, parts of the world and, in a global perspective, must be considered inevitable. T h e only red e e m i n g feature of the destruction of wetlands for p o n d construction is that it is probably less harmfui than dredge-and-fill-type destruction. Ponds may still p e r f o r m a few o f the many roles played by wetlands. Carbon fixation and the transport o f organic material still occurs, but the rate is uncontrolled. Obviously, the ecological balance of estuarine systems is disrupted by conversion of wetlands to shrimp ponds. Shrimp farms tend to e x p a n d into areas that are u n d e v e l o p e d or are currently developed for less lucrative industries, rice farming being a prime example on a global basis. No discussion about the impact of shrimp farms on the adjacent marine ecosystems could be complete without m e n t i o n i n g that the relative impacts of alternative types of coastal d e v e l o p m e n t can, in some instances, be even Inore severe. ESTUARY EUTROPIIICATION FROM StIRIMP POND EFFLUENT
A second i m p o r t a n t effect of shrimp farming on the estuarine e n v i r o n m e n t is the discharge of nutrient-laden p o n d water and eutrophication, or at least hypernutrification, of the receiving body. Under this category, interrelated environmental impacts can take several forms including dissolved oxygen reduction as a resuh o f the, biochemical oxygen d e m a n d of effluent, and increases in phytoplankton, macrophyte, and microbial a b u n d a n c e in response to nutrient inputs. Virtually all shrimp farms use water exchange to some degree (I,ee and Wickins 1992). Phillips et al. (]991) note that p r o d u c t i o n of a metric ton of shrimp uses 16,000 metric tons, 36,000 metric tons, and 55,000 metric tons of water for extensive, semiintensive, and intensive systems respectively. These were the highest water use values r e p o r t e d for the various aquaculture crops considered. In a shrimp farming industry survey, Hopkins and Villal6n (1992) r e p o r t e d an average of 86,000 metric tons of water used to p r o d u c e a metric ton of shrimp, with a slight trend towards less water use in m o r e intensive systems. Phillips et al. (1991) note that the large differences in water c o n s u m p t i o n between species and system types c a n n o t be corn-
pletely explained by interspecific differences in physiological tolerances and suggest that there is r o o m for m u c h i m p r o v e m e n t in water management. As discussed in the next section, this is substantiated by recent research on shrimp-pond water m a n a g e m e n t (Hopkins et al. 1993, 1995). Shrimp p o n d effluent is high in suspended solids, particulate and dissolved organic and inorganic nutrients, and biochemical oxygen d e m a n d (Hopkins et al. 1988, 1993, 1995; P r u d e r 1992; Ziem a n n et al. 1990, 1992). Although the shrimp p o n d may be t e r m e d heterotrophic, with respiration exceeding photosynthesis, the actual dissolved oxygen concentration will be high as supl)lemental aeration is used to protect the crop. For example, in an intensive shrimp p o n d where 25% of the p o n d volume was e x c h a n g e d daily, Hopkins et al. (1993) r e p o r t e d an average dissolved oxygen concentration of 5.4 mg 1 1 at dawn, with saturated or supersaturated conditions in tile afternoon. I Iowever, BOD in the p o n d averaged 8.5 mg 1 l compared to 1.5 mg 1-1 tor the adjacent estuary. Theref o r e , w h e n p o n d w a t e r is d i s c h a r g e d to the adjacent estuary, a reduction in oxygen concentration will occur. T h e degree to which oxygen depletion occurs in the receiving body is a fimction of the quality of the eftluent, decay rates, and the rate of effluent dilution in the receiving stream (Brune 1990). T h e nutrient c o n t e n t of shrimp p o n d water is d e t e r m i n e d by feed and fertilizer inputs and water exchange. IIopkins et al. (1993b) r e p o r t e d the Kjeldahl nitrogen concentration of estuarine water, of an intensive p o n d with 25% daily water exchange, and of an intensive p o n d with 2.5% daily water exchange to be 4.2 mg 1 1, 6.5 nag 1-l, and 9.2 mg 1 ~, respectively. While the concentration of nitrogen and o t h e r pollutants was higher in the 2.5% d -1 exchange pond, the total release of nutrients was actually lower due to in-pond digestion processes. Ziemann et al. (1992) r e p o r t e d 0.7 mg 1-~ total nitrogen in shrimp p o n d water but did not provide information on the feeding rate, stocking density, or water exchange. Nutrient e n r i c h m e n t of the adjacent estuary by shrimp point effluents can be e x p e c t e d to result in increased phytoplankton a n d / o r macrophyte abundance. Shrimp p o n d effluent has elevated levels of nutrients and phytoplankton (IIopkins et al. 1988, 1993, 1995; Ziemann et al. 1990, 1992; Pruder 1992). Thus, the p o n d provides the nutrients for phytoplankton a b u n d a n c e increases in the estuary, and a culture inoculum as well. However, empirical data c o m p a r i n g phytoplankton concentrations in estuarine areas before and "after the discharge of shrimp p o n d effluent are not available. For comparison, no measurable effects on phyto-
EnvironmentalImpactsof Shrimp Farming
plankton were f o u n d associated with net-pen prod u c t i o n o f s a l m o n i d s (Weston 1991). Twilley (1989) n o t e d that red tides have recently been observed in estuaries a r o u n d shrimp farms in Ecuad o r but cited r u n o f f from terrestrial agriculture and estuarine disposal o f sewage t h r o u g h rapid urban expansion as major contributors to n u t r i e n t loading, which may cause such blooms. Since phytoplankton populations require about a day to double in n u m b e r in response to increased nutrient c o n c e n t r a t k m , the shrimp p o n d effluent is probably dispersed too rapidly to allow changes in phytoplankton populations to be easily quantified, except as they relate to direct dilution of phytoplankton in the effluent water. Increased abundance o f sessile macrophytes may be easier to quantit~r as the effect will be m o r e p r o n o u n c e d immediately adjacent to the point of discharge. Hopkins et al. (1993) f o u n d that the suspended solids averaged in excess o f 180 mg 1-1 in an intensive p o n d where 25% of the water was being exchanged each day, although this concentration was only slightly h i g h e r than the adjacent estuary which had high natural sediment turbidity. Susp e n d e d solids in a low exchange p o n d were about twice as high as that o f the adjacent estuary (I Iopkins et ai. 1995). In Hawaii where coastal waters are oligotrophic, the suspended solids load of shrimp p o n d effluent was 50 to 80 times higher than the receiving waters (Ziemann et al. 1992). We must assume that shrimp farm effluent has some effect on microbial activity in the surrounding estuary. T h e ultimate digestion of pollutants from shrimp p o n d water is likely driven by microbial processes. Untbrtunately, the only data available relate to microorganisms of human-health concern. Vibrio vuhzificus, a h u m a n pathogen, was m o n i t o r e d in oysters held in both shrimp ponds and the estuary. T h e bacteria disappeared from oysters within 30 d of being placed in the shrimp pond, while the bacteria concentration increased in oysters held in the estuary (Burrell et al. 1991). Fecal coliform bacteria, an indicator species used to m o n i t o r molluscan shellfish sanitation, are typically no higher in shrimp p o n d s than the estuary (unpublished data, Waddell Maricuhure Center). In recent years, there have been reports of declining growth and increased mortality in shrimp ponds as the n u m b e r of shrimp farms built in close proximity increases (Csavas 1990); examples include Taiwan (Liao 1992) and n o r t h e r n Thailand (Boyd and Musig 1992; Rosenberry 1992). Degradation of water quality" in the adjacent estuaries is cited as the probable cause. In heavily developed shrimp thrming areas such as Thailand, Taiwan, and Ecuador, m u c h of the water being p u m p e d into ponds for water exchange was only recently
29
discharged from o t h e r ponds. In the Guayas River area o f Ecuador, the c o m b i n e d p u m p i n g rate of shrimp farms exceeds the river discharge during periods o f low flow (Twilley 1989). However, the cause of p o o r shrimp p r o d u c t i o n c a n n o t be attributed to any one water quality p a r a m e t e r or even water quality in general (Boyd and Musig 1992). Diseases are also implicated in shrimp p o n d production declines over b r o a d areas. It is possible the disease problems are associated with a general degradation of the s u r r o u n d i n g estuarine environm e n t (Overstreet 1988; Csavas 1990; Pillay 1992). T h e impact of shrimp p o n d effluents on estuarine environments is m u c h m o r e difficult to quantify than wetland destruction t h r o u g h mangrove clearing. Some. baseline surveys necessary to docu m e n t the extent of wetland loss are likely to be f o u n d in the archives o f even the least-developed countries. However, reliable baseline data on estuarine water quality are hard to find even in the United States. T h e problem is exacerbated by the rapidly fluctuating conditions i n h e r e n t in estuarine environments. T h e r e are no quantified reports of biological or chemical changes that have o c c u r r e d in estuarine systems as a direct result of shrimp farming activity. In some cases, the effects have b e e n m o d e l e d u s i n g f i r s t - o r d e r kinetics (Brune 1990) but never verified t h r o u g h field sampiing. It should be n o t e d that density and species richness in the estuary may increase in the immediate vicinity of an aquaculture site because o f increased availability of food resources (Weston 1991). Such effects are well d o c u m e n t e d for various types of fish culture but have not b e e n addressed in relation to shrimp farming. IMPACFS ON INDIGENOUS SHRIMP STOCKS
In most parts of the world, shrimp farming still relies on the capture of wild shrimp. This includes the capture of postlarvae for stocking ponds and the capture of adult broodstock to support hatchery operations. In o t h e r areas, shrimp culture businesses are based on i n t r o d u c e d n o n i n d i g e n o u s species. T h e impacts o f these activities are difficult to assess and have not been well d o c u m e n t e d . Wild postlarvae are collected for stocking ponds a r o u n d less developed extensive farms in Asia, but the practice is perhaps most prevalent in the Penaeus vannanwi and P. stylirostris thrming industries of Ecuador and o t h e r Pacific coast areas o f I a t i n America. T h e r e are n u m e r o u s shrimp hatcheries in this part of the world, but many reduce their o u t p u t when there are a b u n d a n t wild postlarvae to be caught as the wild stock is less expensive and is t h o u g h t to p e r f o r m better in grow-out ponds. Postlarvae are captured by artisan fishermen with sim-
30
J.S. Hopkins et al.
pie fishing devices. T h e r e is probably little envir o n m e n t a l impact f r o m their activity except for the removal o f large n u m b e r s o f juvenile shrimp from estuaries. Fisheries m a n a g e m e n t in the areas most affected is rudimentary. However, population studies of species with similar life histories (other ()penthelyca coastal penaeids) indicate that traditional fisheries for market-size shrimp are m o r e affected by available habitat and environmental conditions during the juvenile and subadult stage than by rep r o d u c t i o n and r e c r u i t m e n t (Lam et al. 1989). T h e n u m b e r of shrimp surviving the estuarine-dep e n d e n t phase of the life cycle is not affected as long as a certain low threshold o f r e p r o d u c t i o n (i.e., a Beverton-Holt-type curve) is achieved. This tends to confirm that natural predation on juvenile and subadult shrimp is intense. We can only speculate on how h u m a n competition for these small shrimp influences natural predator-prey relationships. A second impact of shrimp aquaculture on natural shrimp populations is the collection of wild broodstock. Like the collection o f postlarvae, tile effects on estuarine and coastal environments are not well understood. Broodstock are collected for use in shrimp hatcheries where they are spawned, either naturally or t h r o u g h manipulation of the e n d o c r i n e system. T h e resulting offspring are raised to an advanced postlarval stage in i n d o o r tanks before being stocked in grow-out ponds. At the peak of the Taiwanese shrimp farming industry, the price paid for a mature Penaeus monadon female was as m u c h as $1,800 (US) (Chiang and Liao 1985). With such e c o n o m i c incentives, the fishing pressure for wild broodstock b e c a m e extremely intense and spawners were being shipped into Taiwan from all over the Pacific Rim due to an inability to catch sufficient n u m b e r s locally. T h e farming o f P chinensis in China is d e p e n d e n t on collection of large n u m b e r s o f wild spawners. T h e r e is c o n c e r n that this activity impacts fishery stocks (Xin and Sheng 1992), and the shrimp farming industry is developing technology for closed-cycle r e p r o d u c t i o n prior to an anticipated g o v e r n m e n t restriction on the capture of adult spawners for aquaculture. T h e Latin American P. vannamei farming industry also relies on wild spawners, despite the availability of technology for captive r e p r o d u c t i o n o f successive generations. In o t h e r regions are small localized shrimp farming industries that p r o d u c e successive generations of broodstock and postlarvae without relying on animals from the wild (AQUACOP 1984). As domestication d e v e l o p m e n t o f genetically-superior strains progress (Lester 1983), it is reasonable to expect that this will b e c o m e the n o r m , rather than the exception (I~ester and Pante 1992).
T h e business of shrimp farming often involves the transportation o f shrimp livestock from o n e geographical area to another. Wild shrimp postlarvae may be captured in one area and stocked in ponds adjacent to different estuarine systems, and in some cases, adjacent to different oceans. Hatchery-reared postlarvae are even m o r e likely to be shipped to distant ponds for grow-out to market size. Wild and captive-reared adults are shipped to distant hatcheries where they serve as broodstock. M o d e r n air-freight transportation and inexpensive packaging techniques have made it economically feasible to ship live shrimp a r o u n d tile world. In addition, n o n i n d i g e n o u s shrimp species are reared t h r o u g h consecutive generations in hatcheries. Obviously, the intent o f these transfers is to provide livestock for aquaculture. T h e value of the crop provides an ecom)mic impetus to prevent stock fi'om escaping to the wild. Howevei, sonle degree of escape is probably inevitable where ponds and hatcheries are built immediately adjacent to estuaries will connections via drainage canals. Weston (1991) terms the impacts from these sorts of activities "biological p o l l u t i o n " and notes they are potentially the most serious effects of aquaculture. Several ways through which the escape of shrimp from aquaculture facilities could disrupt ecological balances have been put torward, including introduction of a n 0 n i n d i g e n o u s species that becomes established, interspecific breeding between a native species and an i n t r o d u c e d species with disruption of the gene pool of native species, intraspecific b r e e d i n g between a native species and a domesticated line of the same species with consequent narrowing o f the gene pool, and introduction of a n o n i n d i g e n o u s shrimp pathogen which infects local wild stocks. T h e r e are no d o c u m e n t e d incidences of a nonindigenous shrimp species having b e c o m e established in an area due to importations for aquacult u r e ( p e r s o n a l c o m m u n i c a t i o n L. J. Lester, University of H o u s t o n ) , although there is an unc o n f i r m e d r e p o r t of an unsuccessful attempt to establish P. monodon in an area o f the Caribbean to provide wild broodstock for a commercial shrimp h a t c h e r y and r e p r o d u c t i o n o f i n t r o d u c e d P japonicus in an i m p o u n d e d area along the Italian coast. In Mexico and the United States, c o n c e r n has been expressed over the possibility for establishm e n t of the Pacific species P. vannamei in P setiJerus habitat of the Gulf of Mexico and western Atlantic. Ifikewise, there are no r e p o r t e d incidents of an escaped n o n i n d i g e n o u s shrimp having interbred with a native species. Differences in anatomy, courtship, and mating behavior within the gemls Penaeus suggest that the potential for spontaneous hybridization may be extremely limited (see Bray
EnvironmentalImpactsof ShrimpFarming
and I,awrence 1992; Browdy 1992 for reviews). Successful attempts to create interspecific shrimp hybrids in the laboratory have been r e p o r t e d (Lawrence et al. 1984; Lin et al. 1988; Bray et al. 1990), but the few resulting nauplii were very weak. T h e r e is greater potential for adverse impacts from intraspecific breeding between wild and escaped captive stocks. For these impacts to be realized, large n u m b e r s o f animals nmst escape, survive in an unfamiliar environment, compete, and b r e e d successfully with wild stocks (CATOMA 1992). It is doubtful that intraspecific breeding between a natiw," species and a domesticated line of the same species has o c c u r r e d as there are presently few captive breeding populations that have r e m a i n e d isolated for m o r e than a few generations. T h e r e are u n d o c u m e n t e d reports of two populations o f P. vannanu,~ that were isolated for five to seven generations in tim United States before being destroyed due to virus infections. A population of P. stylirost~is was r e p o r t e d to have p r o d u c e d 10 successive generations in New Caledonia (AQUACOP 1984; Ottogall et al. 1988). At present, these captive populations are well removed from the natural range of the species. Sbordoni et al. (1987) f o u n d a reduction in the level of genetic variability from the F l to the F., generation o f hatchery-produced stocks of P. japonicus as the result of inbreeding depression and a progressive reduction of mean hatch rates. S u n d e n and Davis (1991) rep o r t e d that aquaculture populations of P. vannamei showed slightly lower levels o f heterozygosity and possessed fewer rare alleles than did wild populations. A n o t h e r aspect of this issue is the benefit and threats created by programs that e n h a n c e wild shrimp populations with hatchery-reared livestock. T h e r e is a shrimp e n h a n c e m e n t program for P. japonicus in Japan that, in 1983, released 302 million postlarvae (Uno 1985). In China, shrimp stock enh a n c e m e n t has been in progress since 1983, with 180 million postlarvae released in 1988 alone. Fishery data indicate that the stocked shrimp grow and migrate with a survival o f 32% to 36%. Shrimp stocks in selected bays have increased by 4.5-fold in some cases (Liu et al. 1991). In South Carolina, tile overwintering population o f adult P. set~erus are occasionally killed by abnormally low winter temperatures in the estuary, resulting in r e d u c e d commercial shrimp landings the tollowing year (I,am et ai. 1989). T h e feasibility of overwintering large n u m b e r s of P. setiferus in thermally-enriched ponds to provide a r e p r o d u c i n g population after severe winters may be economically viable and has been the subject of directed research (unpublished data, Waddell Mariculture Center). These sorts of" fishery-related activities pose some of the
31
same biological pollution dangers associated with shrimp aquaculture. Shrimp pathology is being studied in detail for aquaculture systems. While shrimp diseases play an i m p o r t a n t role in the shrimp farming industry, there are no reports of adverse consequences from shrimp diseases having been spread from culture systems to wild stocks. Csavas (1990) r e p o r t e d that instances where disease has been implicated as the cause o f a major shrimp farming epizootic have involved only indigenous diseases. Howevel, in Hawaii, an i n d o o r "super-intensive" shrimp farming project with completely controlled environmental conditions and p r o d u c i n g n o n i n d i g e n o u s P. stylirostris was decimated by infections o f the nonindigenous infectious h y p o d e r m a l and hematopoietic necrosis (IHHN) virus. More effort must be directed to the study o f disease prevalence in wild populations, like the Pacific" coast I H H N survey work by I , o ~ and Overstreet (1991). Pathologists fix)m o t h e r animal h u s b a n d r y industries, such as poultry and swine, believe it is m o r e likely that wild stocks will transmit a disease to captive stocks than the reverse (personal communication, T. Eleazar, Clemson University). In addition, many of these pathogens are diseases o f c o n f i n e m e n t that have little effect on less-stressed wild populations but are epizootic in confined culture populations (CATOMA 1992). O f particular c o n c e r n to shrimp aquaculturists are a suite of shrimp viruses that have been shown to result in catastrophic losses on farms ( I J g h t n e r et al. 1991; I J g h t n e r 1992). Shrimp farmers have d o n e a p o o r j o b of preventing transfers of these viruses from one shrimp farm to a n o t h e r (Colorni et al. 1982). This is due as m u c h to ignorance and a general lack of disease diagnostic capabilities as negligence. Most of the viruses were not described until they had reached epidemic proportions in shrimp farms, and many of the viruses c a n n o t be d i a g n o s e d e x c e p t in a few s p e c i a l l y - e q u i p p e d shrimp disease research centers. WATER USE AND ENTILAINMFNT OF ESTUARINE BIOTA
Shrimp fhrms are largely d e p e n d e n t u p o n the use of estuarine water for filling ponds and for subsequent water exchange. T h e discharge of water and added nutrients, solids, and phytoplankton associated with p o n d communities is a c o n c e r n and so is the removal of estuarine biota in the process of transferring water into ponds. While it is conceivable that e s t u a r i n e o r g a n i s m s c o u l d be p u m p e d into ponds and later discharged back to the estuary with e.ffluent, this is not likely to occur and most organisms entrained will perish in the process. Tile magnitude of the p r o b l e m is exem-
32
J.S. Hopkins et al.
plified in the Guayas River area of Ecuador where the c o m b i n e d p u m p i n g rate o f shrimp farms exceeds the river discharge during periods of low flow (Twilley 1989). While quantitative information is not available, similar situations may exist in certain areas o f Thailand and o t h e r congested shrimp farming regions. N u m e r o u s estimates of the magnitude of entrainment mortality have been m a d e for power plant cooling water intakes (e.g., (;lark and Brownwell 1973; Boreman et ai. 1981; J e n s e n 1982), an environmental c o n c e r n comparable to p u m p i n g large volumes o f water from estuaries for shrimp culture. These estimates are obtained using information on the hydrodynamic characteristics of the adjacent water body, life history of entrained biota, stock size, susceptibility of individual organisms to e n t r a i n m e n t and e n t r a i n m e n t lnortality. Entrainm e n t loss is expressed as the fraction o1 the initial population that would have been killed during a specific time period if no o t h e r sources of mortality existed. This estimate is generally t e r m e d the conditional e n t r a i n m e n t mortality rate. Available e n t r a i n m e n t data suggest that facilities in areas where e n t r a i n m e n t life stages are concentrated and a large p r o p o r t i o n o f the adjacent water body is withdrawn, e n t r a i n m e n t losses are a potential threat to some fish populations (Barnthouse et al. 1984; Holland et al. 1986). T h e long-term consequences of e n t r a i n m e n t are, however, poorly understood because of the lack of understanding about processes such as the degree to which indigenous populations c o m p e n s a t e tbr e n t r a i n m e n t losses (Barnthouse et al. 1984). Even in the absence of an understanding o f the long-term consequences, e n t r a i n m e n t models are useful for identifying vulnerable populations, ranking ent r a i n m e n t relative to o t h e r sources of loss, and evaluating alternative strategies. Power plant e n t r a i n m e n t models are suitable for estimating e n t r a i n m e n t losses due to shrimp culture operations. Unfortunately, the data required to apply these models to estuaries a r o u n d high concentrations of shrimp farms are not usually available. Using the total p u m p i n g rates provided by Twilley (1989) tot shrimp farms a r o u n d the Guayas estuary in Ecuador and the e n t r a i n m e n t model of Goodyear (1977), the percentage impact on planktonic biota can be very roughly estimated by dividing 40,000 by the area (ha) of associated estuary. E n t r a i n m e n t of fish eggs, larvae, and pelagic juveniles o f potential shrimp predators is an important c o n c e r n for the shrimp t~armer (Lunz and Bearden 1963). Since tim scrc'ens n e e d e d to remove fish eggs for effective p r e d a t o r control are unselective, o t h e r entrained species are destroyed
as well. Typically, the shrimp f a r m e r tries to filter inlet water t h r o u g h a 200-1~m screen to remove fish eggs. T h e contiguration may be flat screens or mesh "socks" and screen design affects the labor r e q u i r e m e n t to keep them cleaned o f filtered material. E n t r a i n m e n t o f planktonic forms is not directed or managed by the shrimp farmer, but to varying degrees, shrimp farms rely on entrainment. Entrained plankton include larval and adult stages of a r t h r o p o d s and larval annelids that are c o n s u m e d by shrimp either directly or after varying degrees o f d e v e l o p m e n t within the pond. T h e impact o f entrained biomass on the shrimp crop varies with the degree of intensification. In intensive systems, prey populations are quickly c o n s u m e d by shrimp and are unable to recover, thus making a m i n o r contribution to overall shrimp biomass production. Hopkins et al. (1988) n o t e d that populations of developing polychaetes and o t h e r benthos in intensive ponds were high early in the season but disappeared abruptly as the shrimp became large e n o u g h to forage in the sediment. In the fourth week after stocking shrimp postlarvae, macrobenthic polychaetes r e a c h e d a density o f 20,415 m 2 and there were 134,416 m -~ meiofauna polychaetes. By the sixth week, there were only 2,106 m '* macropolychaetes and 31 m -2 meiofauna polychaetes. By the eighth week, they had been grazed down to 0.2% of their anaximum density. Meiofauna n e m a t o d e s had a similar fate; populations declined from 531,018 m -z to 51,932 m ~ and finally to almost zero in the fourth, sixth, and eighth weeks after stocking, respectively (Waddell Maric u h u r e Center unpublished data). Maier (1991) r e p o r t e d that the zooplankton in these ponds was d o m i n a t e d by Acartia tonsa, Parvocalanus crassirostris, Pseudodiaptomus coronatus, Oithona colcarva, and, most notably, polychaete trochophores. Abundance varied dramatically with no obvious trends except for polychaete trochophores, whose numbers t e n d e d to decline rapidly during the initial weeks. T h e shrimp f a r m e r is also d e p e n d e n t u p o n ent r a i n m e n t of phytoplankton for an inoculum in dev e l o p m e n t of phytoplankton hlooms. Phytoplankton play an i m p o r t a n t role in n u t r i e n t recycling and natural productivity within the pond. Shrimp ponds result in a net gain of p h o t o p l a n k t o n in the estuary as the n u m b e r s o f cells discharged is dramatically higher than what is e n t r a i n e d (see discussion of phytoplankton in the water quality section). ENVIRONMENTAL EFFECTS OF (.~HEMICALS USEI) IN SIIRIMP FARMING
In recent years, use o f c h e m o t h e r a p e u t i c s in a q u a c u h u r e has b e c o m e a concern. In o t h e r ani-
EnvironmentalImpactsof Shrimp Farming
mal p r o d u c t i o n operations, such as the beef and pork industries, antibiotics are frequently used on a continual basis to prevent disease and e n h a n c e growth (CATOMA 1992). Beveridge et al. (1991) lists a wide range of chemicals used ill certain types of fish culture including therapeutics, vaccines, h o r m o n e s , flesh pigments, anesthetics, disinfectants, and water t r e a t m e n t compounds. They note that chemicals used in system fabrication may also find their way into water. T h e impacts o f these chemicals relate to c o n c e r n tor h u m a n health a n d / o r the estuarine environment. Bell (1992) n o t e d that researchers and the pharmaceutical industry have not kept pace with the expanding shrimp farming industry and, as a result, there are few safe., effective, a n d a p p r o v e d t h e r a p e u t i c agents to manage shrimp diseases. T h e FDA points out that c u r r e n t federal and state fimding is inadequate to lneet the needs of the aquaculture industry and suggests an appropriation to address the problem (Anonymous 1991a). Viruses are often perceived as the greatest disease problem in shrimp culture. Howevm, except for disinfectants used to minilnize the spread o f viral disease, there is little c h e m o t h e r a p e u t i c s can do to resolve tile problem. Thus, virtually all of the c h e m o t h e r a p e u t i c s of c o n c e r n are designed to control bacterial diseases. Methods of administering drugs to the shrimp crop include long-term or indefinite baths, dips or short-term baths, and medicated feed. Bell (1992) suggests information be available in four requisite areas before tile use of a shrimp farming c h e m o t h e r a p e u t i c is allowed. T h e four areas are efficacy (Is the c o m p o u n d effective against the disease agent?), animal safety ((;an the desired results be obtained without jeopardizing the health of the shrimp crop?), h u m a n safety (Will there be unsafe residues o f file drug left in edible tissue and does administering the drug e n d a n g e r the aquacuhurist?), and environmental safety (Will the drug make its way into tile e n v i r o n m e n t at harmful levels?). To insure that there is no environmental damage, a d r u g must degrade rapidly within the cultured animal and have a short half-life in the culture system. Tllere should be no discharge during the time the d r u g is active. T h e use o f therapeutics ill shrimp hatcheries d i f fers dramatically from that of shrimp grow-out p o n d s . B o o n y a r a t p a l i n (1990) d e s c r i b e d Asian shrimp hatchery diseases and treatments that included use o f chloramphenicol, c o p p e r sulfate, EDTA, fin-anacc, fiarazolidonc, nitroturazonc, oxytetracycline, trellan, and zeolite. C o p p e r sulfate, EDTA, and zeolite are chemicals used to treat the c u h u r e water rather than the shrimp thenlselves. C a r p e n t e r (1992) provides the results of a world-
33
wide survey of shrimp farms c o n c e r n i n g their use of chemotherapeutics. Shrimp hatcheries r e p o r t e d tile most serious disease problems were viruses. In the Far East, P. monodon-type baculovirus (MBV) was the primary c o n c e r n while western world hatcheries were p r e o c c u p i e d with Baculovirus penaei (BP). T h e only treatments used were an iodine wash for newly hatched nauplii to prevent MBV virus from being inta'oduced with parental fecal material and complete facility disinfection with sodium hyperchlorite (commercial bleach) for BP. Industry respondents from both sides of the Pacific r e p o r t e d problems with luminescent bacteria. In Asia, bacteria are treated with antibiotics such as oxytetracycline, furazolidone, and sulfa-trimethoprim. In the west, bacteria are eliminated by dipping nauplii in a dilute iodine solution before stocking, treating the water with ozone or ultraviolet radiation, increasing the water exchange, and if necessary, treating with antibiotics based on sensitivity. A symptomatic breakdown o f the cells lining the hepatopancreas is treated with antibiotics ill both the east and the west. Many shrimp hatcheries discharge effluent directly into the ocean. A better practice might be to discharge to holding ponds where antibiotics could degrade and deactivate. R e f e r r i n g to Asian s h r i m p f a r m i n g , Csavas (1990) warned that antibiotics are widcly used and misused to combat opportunistic pathogens such as Vibrio species. Indiscriminate use o f antibiotics is a h u m a n health c o n c e r n as it may lead to dev e l o p m e n t and spread o f antibiotic resistance. Unfortunately, there are unscrupulous commercial agents t h r o u g h o u t Asia who persuade u n e d u c a t e d shrimp farmers to use unwarranted and excessive amounts of antibiotics and o t h e r pharmaceutical products. However, in an industry survey of diseases afflicting shrirnp grow WIn\AN. 1995. The effect of low-rate sand filtration coupled with carethl feed management on effluent quality, pond water quality, and production of intensive shrimp ponds. Estuaries 18:116-123.
Hoe~aNs, j. s., R. 1). lla~ttLm.~, e. A. SANnWER,C. 1,. BROWDX, AND A. D. STOKES. 1993. Effect of water exchange rates on production, water quality, effluent characteristics, anti nitrogen budgets of intensive shrimp ponds. Journal of the World Aquaculture Societ~ 24: 304-320. 11OPKINS,J. S., P. A. SANDIFER,R. D. HAMILTON,AND C. L. BRowm'. 1994. Sludge utanagement in intensive pond culture of shrimp: Effect of management regime on water quality, sludge characteristics, nitrogen extinction, and shrimp production. Aquacultural Engineering 13:11-30. I-IoPKtNS,J. S., E A SANnIFER,A. D. STOKES, AND C. 1. BROWDY. 1991. The effect of minimal water exchange on the water quality and production of intensive marine shrimp ponds, p. 33. Program and Abstracts, 22nd Annual Conference, World Aquaculture Society, Baton Rouge, Louisiana. HoPKIXS,J. S. ANDJ. Vn.l,U.Oy. 1992. Synopsis of industrial panel input on shrimp pond management, lnJ. A. Wyban (ed.), Proceedings of the Special Session on Shrimp Farming. World Aquaculture Society, Baton Rouge, Louisiana. JENSEN, A. L. 1982. Impact of a once-through cooling system on the yellow perch stock in the western basin of 1,ake Erie. Ecological Modeling 15:127-144. LArd, C. F.,.]'. D. WHITAR~.R,A_'
