Passing the Panda Standard: A TAD Off the Mark?

Abstract

Tilapia, a tropical freshwater fish native to Africa, is an increasingly important global food commodity. The World Wide Fund for Nature (WWF), a major environmental nongovernmental organization, has established stakeholder dialogues to formulate farm certification standards that promote “responsible” culture practices. As a preface to its “tilapia aquaculture dialogue,” the WWF for Nature commissioned a review of potential certification issues, later published as a peer-reviewed article. This article contends that both the review and the draft certification standards subsequently developed fail to adequately integrate critical factors governing the relative sustainability of tilapia production and thereby miss more significant issues related to resource-use efficiency and the appropriation of ecosystem space and services. This raises a distinct possibility that subsequent certification will promote intensive systems of tilapia production that are far less ecologically benign than existing widely practiced semi-intensive alternatives. Given the likely future significance of this emergent standard, it is contended that a more holistic approach to certification is essential.

Introduction

Significant and increasing portions of global industrial aquaculture production for internationally traded species may soon come under the umbrella of voluntary certification schemes (Vandergeest 2007). Such schemes operated by organizations, including producer groups, retailers, and nongovernmental organizations (NGO), act through product labelling to guarantee that conditions under which farmed aquatic produce is produced conform to various environmental, ethical, and health-related standards. The presence of an “eco-label” is intended to act a source of information for consumers, enabling consumers to assert choices that favor preferred product attributes, thereby facilitating market-based shifts in consumption toward more desirable forms of production.

Production of tilapia1 has grown rapidly over the last decade. Global output of farmed tilapia doubled between 1997 and 2004 (Food and Agriculture Organization (FAO) 2007) and is anticipated to exceed 3 million tonnes (live weight equivalent) by 2010, if not sooner (Fitzsimmons 2006). During the same period, global trade in tilapia quadrupled in volume and rose 6.5-fold in value as the fish become an internationally traded commodity with major markets in the developed world (Food and Agriculture Organization (FAO) 2007).

Several farm assurance certification schemes for tilapia production have been initiated as a result of these concurrent trends. Prominent among these is the tilapia aquaculture dialogue (TAD), a forum established by the World Wide Fund for Nature (WWF). The TAD process is based on development of a set of principles for “responsible tilapia production,” which form the basis for farm certification standards. The “dialogue” is intended to achieve meaningful and inclusive engagement with a group of stakeholders, which includes large-scale tilapia producers, buyers and retailers, environmental NGOs, and natural scientists professionally involved in aquaculture research. Principles are designed to provide criteria for identifying sustainable production practices and indicators against which to measure them. These, in turn, will form the basis of a set of “better management practices,” adherence to which will ultimately earn a retail product the right to display a label denoting the “responsible” nature in which it was produced (WWF 2007). As with other major certification schemes, it is envisaged that, once implemented, compliance assessment and certification of the resulting standards will be conducted by a third-party organization, the Aquaculture Stewardship Council (WWF 2009).

To assist aquaculture dialogue stakeholders to develop certification standards, the WWF commissioned a review of information on culture methods, possible negative environmental and social impacts, and food safety concerns for a number of cultured aquatic species under consideration for certification (Boyd et al. 2005). Findings relating to tilapia were published in a peer-reviewed article that provided a reference point for the subsequent development of TAD principles (Boyd et al. 2005). Because the TAD is an ongoing process that has not yet reached its final conclusion, it represents something of a “moving target.” In this article, therefore, the authors elected to evaluate the most recent version of the draft standards (which remain under public review and subject to further change) (WWF 2008a), alongside the review (Boyd et al. 2005) (which is a complete document, the central assumptions of which continue, with some notable exceptions, to be reflected in subsequent draft TAD standards). Additional information is drawn from supporting documents published on the TAD Web site as outcomes of TAD meetings. Taking this approach allows the evolution of the TAD process and, along with the logic that shaped it, to be traced. Such thorough evaluation is timely and warranted, because a close reading of both the review and the draft standards reveals alarming omissions and misconceptions of the issue of sustainability as it relates to tilapia production.2

Certification Issues

The original review identified 12 issues for consideration by stakeholders in the certification dialogue. These issues, nearly all of which are technical in nature, are listed in Table 1, along with the relative importance that the review’s authors ascribed to each in terms of potential impacts. The issues and principles established in the draft standards are presented in Table 2. A comparison between the categories in the two tables reveals the derivation of the issues in the latter from the former and a high degree of similarity between the two. Because there is some overlap even among the seven issues addressed by TAD, we divided our analysis into four overarching categories in the discussion that follows. These are the following: fish health, resource use, water quality and the aquatic environment, and user conflicts and wildlife.3

Table 1.

Relative importance ascribed by the review to the issues addressed (Boyd et al. 2005) and their relationship with draft TAD principles (WWF 2008a)

Issues	Relative importance	Incorporated into draft principle
Antibiotic use	M	6
Benthic biodiversity	M	4
Chemical use	L	6
Disease transfer	L	6
Escapees and invasive species	H	2/4
Genetic alteration	H	4
Land and water use	H	2/3
Mortality removal	H	6
Natural resource use (including fish meal)	M	5
Nutrient enrichment and water pollution	H	3
Predator control	H	4
User conflicts	M	1/2

L low, M medium, H high

Table 2.

Framework for draft standards for responsible tilapia aquaculture developed by TAD (WWF 2008a)

No.	Issue	Principle
1	Legal framework	Obey the law and comply with all international, national, and local regulations
2	Farm siting/development	Site farms or expand existing farms to conserve natural habitat and local biodiversity
3	Water quality	Conserve water resources
4	Biodiversity/genetic impacts	Conserve species diversity and wild populations
5	Feed	Use resources efficiently
6	Health management/disease	Manage disease and pests in an environmentally responsible manner
7	Social issues/social responsibility	Be socially responsible

Fish Health

The review (Boyd et al. 2005) notes that tilapia are more resistant to disease than most farmed fish species and that there is little use of antibiotics, drugs, and other chemicals for disease control in their culture. It also suggests that certification programs should discourage the use of antibiotics and drugs and should disallow their use as prophylactics. According to the review, the spread of disease from farm to wild fish is a possibility but is of low relative importance. The review also indicates that a high incidence of disease on a farm can be expected at sites where water-quality parameters are outside optimal ranges and that fish are often susceptible to disease when stocked at high density and subject to stress (Boyd et al. 2005). The draft standard (WWF 2008a) reiterates this position, stating that “there are few recorded cases of disease directly attributed to tilapia farming.” It also replicates the review’s position in advocating the use of mortality rates as a key indicator.

Although it is true that the hardy characteristics of tilapia are some of the features that make them an attractive culture species, evidence emerging from Asia suggests that disease is becoming a significant and growing problem as production expands and intensifies. Very recent evidence from Thailand suggests that, in a number of watersheds, a pathogen previously unknown in tilapia, Microsporidum, has been responsible for extremely high mortalities in cage-raised tilapia and is apparently also fatal to other species of fish (W. Turner pers. comm.). One assessment of tilapia culture in central Thailand found that disease was the second most common reason, after low farm-gate price, for farmers’ failure to break even; 30% of pond-based and 58% of cage-based farms failed to do so as a result of a disease outbreak (Belton et al. 2009).

Streptococcus iniae is one of the most economically significant diseases prevalent in farmed tilapia (Intervet 2006). Antibiotics are only usually effective in treating a bacterial outbreak if treatment is applied very early during the course of the disease, and, in most cases of Streptococcus outbreak, oral antibiotics are ineffective (Shoemaker and Klesius 1997). Persistent antibiotic application is also likely to result in the emergence of resistant strains (Lehane and Rawlin 2000), and concerns regarding food residues create negative consumer perceptions. Standards could be used to compel farmers to adopt nontherapeutic preventative measures; antibacterial vaccines are in rapid development, although without simultaneous improvement in the diagnostic capacity of smaller-scale operations such a standard would simply impose a relatively greater economic burden on this group.

Good management is the best prevention measure for streptococcal disease. This includes maintaining good water-quality parameters, removal of dead fish, good diagnostic capacity, and maintenance of farm records that monitor disease incidence on a production-unit basis. Stocking rates may also influence the health of tilapia, with high densities linked to high disease transmission rates and mortality (Bunch and Bejarano 1997). Transmission of Streptococcus from Nile tilapia to cyprinids has been observed under laboratory conditions (Garcia et al. 2004), but whether reservoirs of disease in cultured tilapia pose any threat to wild fish of other species is open to question.

As indicated above, tilapia produced in cages (floating enclosures with mesh sides) under intensive conditions may be more susceptible to infectious pathogens than those raised in ponds at lower densities (Chinabut 2002). The review also notes that poor water quality related to high stocking densities may cause frequent or constant stress (Boyd et al. 2005). These conditions are often linked to outbreaks of streptococcosis (Bunch and Bejarano 1997). Cages are open to the surrounding environment to facilitate water exchange, which provides stocked fish sufficient oxygen and removes metabolites and uneaten feed. The open nature of cages results in extremely poor biosecurity, with fish readily exposed to pathogens and pollutants from the surrounding environment. A study from Thailand found that failure to break even on at least one occasion was far more common among farmers who operated cages in rivers and canals than among those practicing pond-based culture, with poor performance in the former commonly resulting from mortalities related to industrial, agricultural, or municipal pollution. Every single cage farmer interviewed for the study had lost money at least once, whereas only half of pond farmers (most of whom had farmed for far longer) ever had. However, large fluctuations in market demand and/or value for cage fish compared with pond fish were found to be an equally important reason for incurring losses (Belton et al. 2009).

Resource Use

This category addresses the use of land, water, and other natural resources. The review refers to land use in terms of the surface area required for pond production. It notes that pond culture of tilapia is “land intensive,” in comparison with cage-based operations, which require very little land other than small bank-side staging and storage areas, and suggests that farms should not be built on wetlands or habitat protected by law for conservation purposes, assigning the issue a high level of importance. The current draft standards also reiterate the need to prevent wetland loss (WWF 2008a).

However, the historical codevelopment of pond-based tilapia aquaculture with agriculture and urbanization (Little et al. 2007) means that it is unusual for the construction of inland aquaculture ponds to occur on areas of high conservation value, such as wetlands, the more common route being conversion of agricultural land or exploitation of “borrow pits” dug for house or road construction. Land used directly for pond-based tilapia production, therefore, is typically of limited ecological significance. There is also evidence that pond excavation may lead to the creation of a mosaic of terrestrial and aquatic habitats, thus benefiting aquatic biodiversity (Little et al. 1996). More importantly, the land and space resources occupied by aquaculture operations are of far lesser significance than the ecosystem area or “ghost hectares” required to supply resources that sustain the activity (Beveridge et al. 1994).

The draft TAD standards downplay water use, referring only to a requirement to prevent salinization of groundwater and limit nutrient loadings. This contrasts with the draft Pangasius aquaculture dialogue standards (WWF 2008b) for which water use (consumption) features as an important issue. This framing of the water-use issue obscures the need to link the quality of intake and output water to any net consumption (or production) of water per unit biomass gained during culture. Feed-associated water use, therefore, is overlooked. This may be significant, because intensive fish production systems reliant on commercial fish feeds produced by using primary outputs of terrestrial agricultural production are potentially more water consumptive in this regard than semi-intensive alternatives, because more than 1 m³ of water is required to produce a single kilogram of grain (Pimental et al. 2004). Furthermore, the effects of pond construction are misconstrued because, although subject to losses through evaporation and seepage, ponds tend to be net contributors to overall water budgets since they serve to harvest and store rainwater which would otherwise be lost as run-off (Clemmens et al. 2008).

The only other resource considered by the draft standard is fish meal.4 Use of fish meal derived from marine fisheries for aquaculture may adversely impact the ecosystems from which it originates (Naylor et al. 1998; Naylor et al. 2000). However, because rates of fish meal inclusion in formulated tilapia feeds are relatively low (4–8%), it is theoretically possible to recover greater amounts of fish meal from tilapia processing waste than is used in the feeds on which they are grown. The most recent draft TAD standards develop this concept further, introducing a quantitative measure, the “Inclusive Feed Fish Equivalency Ratio (IFFER)” to “account for how much fish meal and oil is used and how much is produced via the production process” (Boyd et al. 2005). Under this standard, the IFFER must be ≤0.5 and, if >0, then “the origin of fish meal and oil should be from fish stocks that have an average score >7.5 with no individual indicator below 6.0” according to data from “Fishsource,” an online tool hosted by the NGO Sustainable Fisheries Partnership. Such an approach is of merit (provided that the offset is a measurement of the actual quantity of fish meal reclaimed from tilapia carcasses during processing, as opposed to the quantity theoretically possible), in that it effectively stipulates maximum levels of fish meal and oil inclusion in tilapia diets and ensures that they be requisitioned from sustainably managed sources.

There should, however, also be an evaluation of the opportunity cost of using byproducts for fish meal. Flesh–bone separation, mincing, and other techniques provide increased opportunities for additional products for human consumption to be made with higher unit values than fish meal (Connell and Hardy 1982). Furthermore, it is important to acknowledge that, although meal derived from farmed fish, such as tilapia, may be partially substitutable for wild fish in feed, they cannot replace the function of wild fish in the ecosystems from which they were extracted (Jacquet and Pauly 2007). Fish-meal inclusion rates in herbivorous fish feeds are low, and fish protein conversion occurs more efficiently than in other livestock production system. However, only a relatively small portion of tilapia and carps are presently produced by using formulated diets, and trends point toward the ongoing intensification of production of these species throughout Asia (FAO 2006). The sheer volumes produced, particularly when projected into the future, are such that, notwithstanding a lessening the demand for fish meal in poultry and pig farming (Tacon et al. 2006), cumulative impacts on demand for fish meal resulting from their intensified production may ultimately be far greater than at present (Naylor et al. 2000).

The standards fail to take into account either the scale or relative efficiency of natural resource consumption associated with intensified tilapia aquaculture. At present, the majority of global tilapia production may still occur in ponds under semi-intensive management conditions in which pond water is fertilized to stimulate production of phytoplankton and other microorganisms, which are used by tilapia as natural feeds, usually with additional supplemental feeding (Edwards 1993).5 Semi-intensive systems typically use local byproducts and wastes from other human activities as fertilizers and supplemental feeds, and are thus integrated into the agro-ecosystems in which they are located (Little and Edwards 1999).

A vast array of resources are exploited for this purpose: crop processing provides rice bran and oil cakes; slaughterhouses provide entrails, blood, and bone; food manufacturing, breweries, and distilleries provide diverse organic residues; and intensive livestock production provides a ready source of manure for pond fertilization (Edwards 1998). Nutrient utilization is most efficient when tilapia are fed nutritionally complete diets (Edwards 1993), and significantly shorter grow-out times reduce feed energy requirements for stock maintenance as well as increasing labor and capital efficiency. However, broader definitions of efficiency are relevant, because, considered in toto, integrated systems may be more efficient in terms of nutrient and hydrocarbon utilization (Little and Edwards 2003). A significant proportion of total nutrition can also be obtained from natural feed produced in situ, stimulated by the residual fertilization within pellet-fed ponds.

Intensive aquaculture is heavily reliant on fossil fuel energy inputs (Folke 1988) because of the numerous steps involved in production of complete fish feed (production of machines, fertilizers, pesticides, and ships to produce feed ingredients; the subsequent manufacturing processes to which they must be subjected; and the transportation of these ingredients over long distances), which all require combustion of hydrocarbons (Papatryphon et al. 2004). Intensive tilapia culture also possesses a high degree of dependence on external ecosystems, requiring resources from large ecosystem areas outside the farm to produce its feed, to assimilate its nutrient wastes, and to maintain dissolved oxygen (Berg et al. 1996). Feed ingredients used in complete feeds (soy meal, fish meal, maize, wheat, linseed, etc.) are sourced directly from primary agricultural and fisheries production (i.e., cultivated or captured for the sole purpose of inclusion in fish feed) and are traded internationally as commodities and are transported over many thousands of miles. Furthermore, production of soy, which comprises approximately 60–70% of the protein in most commercially available formulated tilapia feeds, has been linked to a variety of negative ecological and social impacts, of which perhaps the most troubling is widespread deforestation in the tropics (Fearnside 2001). Using the TAD as a means to endorse soy products certified with respect to their environmental and social credentials could, therefore, represent an opportunity to mark out tilapia as a highly sustainable “frontier” product. Because the WWF is in the process of certifying soy producers through a separate initiative that is not at present linked with the Aquaculture Dialogues, the failure to do so represents an unfortunate and short-sighted lost opportunity (Little and Belton 2008).

In contrast, the agricultural and agro-industrial byproducts used as inputs for integrated systems require less differentiated ecosystem space to produce, because they come from areas already appropriated for food production for human consumption. Integrated pond culture that uses such nutrients, therefore, increases the quantity of food produced per unit of total ecosystem area (Kautsky et al. 1997). Furthermore, the nutrients on which integrated tilapia production depends require lower hydrocarbon expenditure to prepare or, because they are generally produced within or close to the agro-ecosystems where tilapia farming occurs, transport to the pond (Belton et al. 2009). The ecological footprint associated with production of inputs and assimilation of waste outputs, therefore, is far larger for intensive cage farming than for intensive or semi-intensive pond farming (Berg et al. 1996).

Water Quality and the Aquatic Environment

In relation to nutrient enrichment and water pollution, the initial review concluded that production of tilapia in ponds where little water exchange occurs allows natural biological processes time to assimilate much of the nutrient originating from the application of feed and fertilizers, whereas cages, in which these processes cannot occur, have greater pollution potential. Solids (primarily uneaten feed and feces, and their decomposition products) and soluble nutrients are discharged continuously from cages. For semi-intensively managed ponds, wastes are mainly discharged during harvest and are composed largely of plankton and clay particles that have a lower oxygen demand than cage-waste products (Boyd et al. 2005).

Benthic effects result from deposition of organically enriched sediments at much higher rates than would occur under natural conditions. This can cause sediments to become anaerobic, altering the composition of communities of benthic organisms. Although elevated sedimentation may occur at the outflows of land-based aquaculture systems, impacts in the vicinity of cages are typically far greater (Boyd et al. 2005). The same applies to nutrient loading: in terms of weight of nitrogen (N) and phosphorus (P) released to the environment for each kg of fish produced, open intensive systems that exchange water with the surrounding environment (cages or raceways) are 7–31 and 3–11 times more polluting than static ponds, respectively (Edwards 1993). More than 80% of N and P inputs to semi-intensive pond-based fish culture are immobilized in sediments on the pond bottom, whereas no nutrients are sequestered by open intensive culture systems, which discharge 73% N and 86% P to the external environment (Edwards 1993). Appropriate harvest and draining methods can reduce the discharge of pollutants from tilapia culture ponds. In addition, whereas pond sediments may also be applied to terrestrial crops as an excellent fertilizer, commercial adoption of technologies for the recovery of wastes from cages used for the on-growing of tilapia have not been developed. However, unlike the review that preceded them, the most recent draft TAD standards make no mention of any difference between cages and ponds. In fact, the TAD Steering Committee (of which several key members are with cage-culture-based enterprises) now holds that all the major culture systems (cages, ponds, and raceway systems) must be considered equal.

The dialogue opts to address rates of eutrophication increase rather than attempting to determine the carrying capacity of receiving waters, because “addressing an impact rather than an indicator dissuades the debates around the ability for systems to assimilate nutrients” (WWF 2008a, p. 6). Therefore, “rather than requiring an assessment of the P carrying capacity of the proposed receiving waters, the TAD is proposing to address the actual level of impact itself—the fluctuations of dissolved oxygen in receiving waters.” Producers will also “be kept to strict limits of chlorophyll a and total phosphorus.” Total farm P output will be calculated as “the amount of P released into the natural environment per mt of fish produced,” with “phosphorus not included in fish at harvest” (WWF 2008a, p. 15) considered the amount of P released into the environment. This logic is understandable but flawed given the difficulties involved in accurately calculating the carrying capacity of receiving water bodies. The draft standards themselves state that “quantifying the amount of phosphorus in effluents is complicated as a result of various feeding times, different times for drain harvests of ponds, precipitation of phosphorus for particular waters, dissolution of phosphorus for specific waters, specific soil phosphorus absorption conditions and the fact that there is no point-source of effluent from cage operations” (WWF 2008a, p. 15), and yet by effectively reducing the equation to P in = P out, the standard fails to account for important site-specific and more generalized qualitative differences in nutrient dynamics and their related effects, thus painting a far more favorable picture of nutrient emissions from open systems than is warranted. The singular attention to P is also misplaced, because, although this element is generally acknowledged to be the primary limiting nutrient in temperate waters, conditions in the tropics are more complex and N is known to be more important (Knud-Hansen et al. 2003).

There are further inconsistencies. The draft standards take the position that it is impossible to achieve zero nutrient impacts to receiving waters with commercial production systems (WWF 2008a). This is incorrect. “Zero impact” is possible in recirculation systems (RAS), because nutrient wastes can be completely retained as a concentrated byproduct suitable for strategic reuse. RAS can also be very conservative in terms of water use, and modern designs prioritize efficient energy use. Investment in RAS for tilapia production is increasing in Europe and North America in response to niche market demand for locally produced food (Little et al. 2008). Bizarrely, however, RAS systems are discounted from certification under the current draft TAD standards, which consider them to be a “trade barrier to small scale farmers,” despite “not having enough volume to shift global markets” (WWF 2008a, p. 7).

User Conflicts and Wildlife

The review commissioned by the WWF (Boyd et al. 2005) reports that, although there are few documented instances of tilapia farms depriving local communities of traditional privileges, some evidence of user conflicts exists and is worthy of further exploration by certifiers. The draft TAD standards differ somewhat, however, by introducing a separate social standard. User conflicts are not mentioned in this standard, and the only “social” issues incorporated relate to labor regulations and worker rights (WWF 2008a). This would appear to favor large vertically integrated enterprises (the dominant voice among the commercial representatives of the steering committee), although being less well suited to addressing the more informal labor relations within medium scale and household enterprises more typical in Asia.

The review’s observation that piscivorous birds and other predators are sometimes killed by fish farmers to prevent fish predation and that certification should require nonlethal predator control (Boyd et al. 2005) has been adopted in the draft standard. It is questionable whether this issue deserves the high relative importance ascribed in Table 1. Avian predation is primarily an issue when tilapia are juveniles and can be minimized, without harm to the birds, through the use and correct management of the right type of visible netting over nursery ponds (Nemtzov and Olsvig-Whittaker 2004). A more proactive approach might encourage the setting aside of water resources and associated land as dedicated habitat to actively encourage the presence of wildlife compatible with fish culture and to mitigate any impacts on wildlife associated with farm construction. The creation of artificial wetlands for this purpose could also serve an additional bioremediation function by sequestering nutrients discharged from ponds during harvest.

Discussion

Prioritizing Sustainability Issues

When the TAD was set up in 2005, its initial focus was to provide certification for the developed world’s largest market for tilapia, North America. The majority of fresh tilapia sold in North America originates from Central America, with far larger volumes of frozen imports that originate from China and Taiwan (Food and Agriculture Organization (FAO) 2007). Central America is unusual in that most farms are strongly export-oriented, corporately owned, industrial in scale, and intensively managed. This contrasts with Asia, where 78% of global tilapia production takes place and the majority is produced less intensively. The operations of most, if not all, the tilapia producers engaged as stakeholders in TAD are located in Central America. The prioritization of issues arrived at in the initial review (Table 1) and still largely reflected in the draft standard, therefore, reflects an orientation toward certification of an atypical and, in absolute-volume terms, far less significant production system.

When integrated semi-intensive culture systems are also considered, it is clear that by far the most important sustainability issues revolve around efficiency of resource use (Belton et al. 2009). Intensive culture methods demand a high throughput of matter and energy, the hidden costs of which are reflected in a sizeable ecological footprint. This contrasts sharply with integrated culture methods that enhance efficiency of natural resource consumption and, as a result, embody a high degree of sustainability (Kautsky et al. 1997; Daly 1999). Standards for tilapia production, such as those likely to result from TAD, therefore, may do little to improve the sustainability of the industry as a whole, although the potential to generate some localized farm-level improvements among participating producers exists. The formulation of standards in this manner (i.e., treating the localized ecological impacts of individual farms as the main focus of environmental degradation and neglecting the important negative externalities associated with the production process) represents a serious failure to grasp the significance of ecological impacts occurring beyond the immediate vicinity of the farm. It, therefore, is necessary to seriously question the validity of any standards for tilapia that equate responsible production practices solely with highly resource consumptive intensive culture systems.

The Comparative Sustainability of Cage- and Pond-Based Tilapia Culture

Although significant nutrient losses may occur in ponds through seepage (Muendo et al. 2005) and, periodically, when water is exchanged or drained, cages discharge a far greater proportion of nutrients into receiving waters than do semi-intensively managed ponds (Edwards 1993) and are thus, in general terms, more likely than ponds to promote or add to the eutrophication of surrounding water bodies. This may alter the makeup of ecological communities and their function and, in the most severe cases, result in oxygen depletion that kills fish and other aquatic organisms (Gong and Zie 2001; Helminen et al. 2000). Escaped fish should also be considered wastes (Beveridge et al. 1994), and, although ponds are by no means sufficiently secure to preclude the possibility of fish escaping, the likelihood of escapes is considerably lower than from cages. These factors mean that cage culture is in essence subsidized (the negative externalities associated with the wastes it produces being borne by receiving ecosystems), whereas, in intensive or semi-intensive pond culture, a large proportion of nutrients are recycled internally or are retained in bottom sediments, which remain on-farm and can potentially be applied to terrestrial crops as a fertilizer. This practice remains rare, but the increasing real costs of inorganic fertilization are likely to stimulate greater efficiency through such strategies (Karim 2006).

Extremely high stocking densities in cages place fish under stress, rendering them potentially more vulnerable to the incidence of infectious disease than they might be if stocked in ponds at lower densities (Montero et al. 1999; Snieszko 1974). This increases the likelihood that medication will be applied and more readily released to the wider environment than if used within ponds. Fish raised in cages are also highly vulnerable to external sources of pollution. In addition to impacting the resilience of smaller producer livelihoods, this has possible implications for food safety and the multifunctionality of water bodies in which cage culture is practiced. The potential for disease transfer from cage fish to wild populations of tilapia or, conceivably, other species is also high because of their openness and high rates of water exchange. Because they are usually located in water bodies with multiple users, cage-based aquaculture also may have a greater potential to cause social conflicts than farming in ponds constructed on privately owned land, notwithstanding issues of access, ownership, or tenure that might be relevant in terrestrial farming.

Although these factors do not make cage-based tilapia culture de facto unsustainable, because careful siting can ensure that the assimilative capacity of the receiving water body is not exceeded, they do tend to make it relatively unsustainable in comparison with land-based systems. Potential problems associated with cage culture may be compounded by production in waters relatively free of anthropogenic impacts. Several Latin American-based export-oriented producers, some of the most influential stakeholders in TAD, specifically emphasize the “pristine” environments in which their culture systems are located as an attribute in the marketing of their product. Tilapia are naturally adapted to grow in eutrophic conditions and can perform well in water that is somewhat degraded and would be unsuitable for other species (Lowe-McConnell 2006). The draft TAD standards specifically state that pristine water will be protected by “limiting the amount of impact” (WWF 2008a, p. 6). There seems to be little equivalence to this position when compared with the conservation of other highly limited pristine ecosystems (rain forest, coral reefs, etc.), and the adoption of such by the WWF would appear untenable. It, therefore, is by no means “responsible” for any certifying body to endorse cage culture of tilapia in relatively intact ecosystems when it can be practiced in locations already subject to anthropogenic degradation.6

Furthermore, in deep stratified lakes and reservoirs (where much of the Latin American cage culture referred to above takes place), there is the potential for cage wastes to accumulate in the hypolimnion over long periods, resulting in anoxic conditions, which only become apparent when low water levels or changing weather conditions cause sudden mixing throughout the water column. This phenomenon has been documented as resulting in severe mortality among both cage and wild fish (Abery et al. 2005; Asian Development Bank (ADB) 2005). However, Northern consumer understandings of what constitutes clean food production are such that production in “pristine” waters is confused with desirable “natural” attributes, whereas fish production under less aesthetically appealing conditions is likely to be perceived as less acceptable. Certification standards intended to improve the industry’s ecological sustainability should not reinforce this misconception.

Intensive cage culture has been identified as being the only solution to meeting the increasing demand for commodity fish, based on limited land area being available for conversion to ponds (Delgado et al. 2003). However, the recent rapid scale-up in freshwater culture of Pangasius catfish culture in Vietnam and of carps, tilapia, and catfish in Bangladesh, Myanmar, and India suggests that this is far from the case. It has been suggested that any specific advantages of ponds compared with cages for commodity-scale production of tilapia have declined; in general, the relative importance of production compared with costs associated with other parts of the market chain diminish in such globally traded products. Thus, the importance of culture systems (feed and management) in the overall price to a Western consumer for tilapia compared with distribution, marketing costs, import export duties, etc., is relatively low (Dempster 2007). The WWF needs to consider the likely patterns of tilapia trade into the future, however, and acknowledge that most consumption will continue to occur in Asia close to production and that under these conditions retention of water and nutrient efficient intensive or semi-intensive production is more sustainable.

A review of certification issues in tilapia culture commissioned by the WWF in 2004 (Boyd 2004), which preceded the later multispecies review, referred to throughout this article (Boyd et al. 2005), adopts a position that is similar in certain respects to that which we advance here. This first review (Boyd 2004), which ranks production systems in “order of environmental friendliness,” lists cages as 6 of a possible 7 culture options (p. 1), noting that “certification possibly should be denied to some production systems which have a high potential for causing water pollution, e.g., cages and net pens” (p. 2), in part because they “tend to release much larger amounts of waste in effluent per unit of production than pond culture systems” (p. 26). Conversely, the reviewer observes that “if managed properly, much of the waste from tilapia propagation will be assimilated in ponds,” that “ponds managed for semi-intensive production have better quality water than those managed for intensive production” (p. 26), and that “concentrations of potential pollutants increase as production intensity increases” (p. 18). The review also prioritizes nutrient reuse as a key indicator of good practice, ranking “raceways and cages integrated into irrigation systems” as the most “environmentally friendly” culture option, on the basis that “production systems integrated with irrigation systems, do not cause pollution and should be prime candidates for trial certification” (p. 1). Pond-based greenwater culture systems are also potential candidates for certification, given that “commercial fertilizers can be used in aquaculture without causing adverse environmental impact” and incorporating guidelines on fertilizer use into certification programs is recommended (p. 26). Given that this orientation of values has subsequently shifted so radically toward a position that legitimizes, de facto, intensive, nonintegrated, open-culture systems as prime candidates for certification and pays scant regard to other options, it is reasonable to argue that the TAD has undergone industry capture, becoming fundamentally compromised in the process.

Product Quality

At present, tilapia produced in ecologically sustainable semi-intensive and intensified pond-based culture systems are consumed mainly in Asian domestic markets, where cultural preferences and expectations relating to food fish differ significantly from those in the developed world. Tilapia from these systems are generally harvested at sizes smaller than those preferred in Northern markets, but experience from countries, e.g., Thailand, in which affluent urban consumers increasingly demand bigger fish, demonstrates that large tilapia can be produced in large-scale moderately intensified systems that retain many hallmarks, and benefits, of integrated semi-intensive culture (Belton et al. 2009).

A more significant objection to semi-intensive tilapia production for Northern markets relates to the sensory quality of fish from fertilized systems. Whereas, musty “off-flavor” in tilapia caused by the absorption of geosmin and 2-methylisoborneol (two metabolites of cyanobacteria that often occur in surface water) may be acceptable to many consumers in Asia, it typically signifies inferior quality product to consumers in export markets (Tucker 2000). However, raising fish in unfertilized systems by using formulated feeds is by no means sufficient to guarantee high sensory quality (Eves et al. 1995); off-flavors are related to many aquaculture systems, including full recycle systems. In any event, the presence of off-flavor producing chemicals in fish tissue can be reduced to below sensory threshold concentrations by holding live fish in clean water without feed for several days postharvest and is a well-established practice in the US channel catfish industry (Tucker 2000). Flesh color is also an issue, because fish raised in semi-intensive systems retain white muscle color less effectively than fish with little or no natural feed in the diet, particularly after prolonged storage; however, carbon monoxide, routinely used for humane slaughter of fish, is used successfully in modified atmospheric packaging to overcome this problem (Mantilla et al. 2008).

Furthermore, results of research indicate that reductions in the natural food produced in situ in tilapia diets through the intensification of feeding regimes results in a progressive negative alteration in ratios of omega-3 and omega-6 fatty acids, rendering them less valuable for human consumption in nutritional terms (Karapanagiotidis et al. 2006). These factors suggest that drivers for intensification justified in terms of superior product quality may be misplaced, particularly when more holistic values are considered.

Conclusion: Setting More Sustainable Standards

Because market acceptance issues pertaining to product size and quality can be managed to ensure compliance with the demands of Northern consumers, certification standards aimed at ensuring sustainable tilapia production should focus on minimizing the management intensity and negative externalities of production systems. It is clear that pond-based culture systems are preferable to cage culture in terms of sustainability. Although cage culture may appear more sustainable if practiced within the carrying capacity of the water body, the risks of uncontrollable contamination, turnover, and other factors make this unlikely long term. More semi-intensive culture practices reduce the need for external inputs by stimulating production of natural feed through fertilization. A case has been made for the promotion of semi-intensive culture based on reducing costs. This cannot be achieved in cages, which by virtue of their openness to the surrounding environment are less biosecure and generally more polluting than static ponds. The commonplace use of livestock manures to fertilize ponds poses minimal risk to human health given appropriate postharvest product handling (Phong Lan et al. 2007), but acceptability of such practices to consumers in the developed world would depend on appropriate pretreatment. Manufactured inorganic fertilizers can also be used for the same purpose within closed pond systems and, despite lacking the benefits of locally sourced manures in terms of nutrient recycling and low hydrocarbon consumption, reduce the need to apply formulated feeds. Feeding regimes can also be designed to optimize nutrient utilization from combined natural and supplemental feeding (Yi and Lin 2002).

Integrated semi-intensive tilapia culture can recycle locally produced nutrient wastes and byproducts from other human activities as feeds. This minimizes the appropriation of ecosystem space and services required to produce feed ingredients, in effect extending the productivity of the areas used in their cultivation. Certification standards should encourage replication of these features as far as is practicable to reduce use of feed ingredients cultivated as primary products and discourage the transport of feeds over long distances. Certification should also prioritize the use of wastes from other fish species that have been processed for human consumption as a source of protein in tilapia diets.

All these observations point to the inadequacy of certification approaches that focus primarily on the localized impacts of the farm alone. They also imply a need for more rigorous certification, which accounts for resource use, negative externalities, and other impacts along the entire value chain for the certified product. At present, although there is ample evidence, as presented in this article, for major differences in relative sustainability between systems of tilapia production at alternative levels of intensity, there has yet been no definitive comparative study of this nature. Obtaining such data should represent a priority for certifiers and would be a crucial first step in the creation of a holistic, vertically integrated certification standard for the entire value chain that goes far beyond the more simplistic “better management practice” guidelines and farm assurance standards currently under formulation.

Any standards introduced to the market should be rational, robust, and capable of withstanding close scrutiny from competitors, consumers, and other interested parties. This is particularly important where, as here, a new benchmark is being established, because it is likely to become the standard against which subsequent standards, possibly by other certification bodies, will be measured. Given the ratchet effect of standards as they develop, as demonstrated by those for organic aquaculture (Aarset et al. 2004), it is vital that their foundations are sound. The potential scope for miscuing consumers is substantial, and the task of communicating accurate messages about the sustainability of fish resources such that they will be perceived correctly by consumers is notoriously difficult. Once communicated, subsequent modifications, which we contend are necessary here, are all the more difficult to convey. Indeed, the WWF has recently demonstrated the potentially contentious nature of such issues with its “Stinky” fish campaign, which was initially intended to promote Marine Stewardship Council labelled fish products but resulted in widespread dissociation from the aquatic food product sector and ultimately its withdrawal. Source credibility remains key to the effectiveness of messages sent, and there is a cost on repeated usage or revision.

This is of particular importance in the case of tilapia. For products such as salmon, production has probably already plateaued and the organization of production and trade stabilized, limiting the variety of options available for certification and the types of messages produced. For tilapia, however, all indications point to large and continuing production increases and an expanding share in Northern markets for aquatic food products well into the future. Thus, it is critical that certification efforts conceptualize sustainability issues convincingly and accurately from the outset if they are to allow consumers to exercise informed choices convergent with discouraging destructive practices and fostering positive ones through the market place. Should TAD fail in this regard, only to be subsequently discredited in the light of greater public scrutiny, it would prove damaging, not only to the WWF and its other certification projects but, potentially, to the concept of eco-labelling as a whole. Thus, there is clearly a need for the WWF to review the TAD from the perspective of broader sustainability, delinking it from the rather narrow perspective of marine ecosystem protection that currently dominates thinking on feed inputs.

Although Asia consumes more than two-thirds of the world’s aquatic produce, very few Asian consumers as yet discriminate among products on the basis of their environmental attributes (Naylor et al. 1998). There has also been limited participation by Asian stakeholders in the TAD process to date. This is, in part, a function of the greater ease with which larger stakeholders and consumers in the Americas can be engaged. In effect, an asymmetric dialogue has resulted, biased toward standards that have less to do with global sustainability and are more reflective of the ability of well-resourced vocal stakeholders to secure beneficial endorsements through active participation, and the global north location of the main public constituency from which support for the big marine conservation orientated NGOs involved in TAD derives. This dislocation provides further cause to query the utility and purpose of such standards.

The TAD approach might arguably represent the first necessary step in an evolving process that becomes more geographically inclusive and responsive to the global realities of production and consumption as it develops, were it not to prejudice future opportunities for pond-based producers in Asia and elsewhere. At present, however, producers who farm tilapia in more sustainable, less-intensive systems (and possess weaker market power than the larger intensive export-oriented producers engaged by TAD) are likely to be most adversely affected by a set of certification standards that emphasize the perception of buyers and consumers that fish raised in such a manner are of poor quality. These producers are unlikely to be able to afford subsequent communications challenges to remedy any incorrect perceptions the market has formed. If the stated ultimate goal of the WWF’s (2008c) Web page: “to build a future where people live in harmony with nature” (Karapanagiotidis et al. 2006) is to be achieved with respect to tilapia, then significant redirection is required. Assumptions on which the current TAD process is founded contain sufficient fundamental flaws to clearly fail this challenge.

Biographies

Ben Belton

is a doctoral candidate at the University of Stirling. His research interests include the political ecology and political economy of aquaculture development.

Francis Murray

is a research fellow at the University of Stirling. His research interests include sustainability and development.

James Young

is a Professor of Applied Marketing at the University of Stirling. His research interests include aquatic food supply chains and their related fish and food markets.

Trevor Telfer

is a Senior Lecturer at the University of Stirling. His research interests include modelling environmental impacts of aquaculture wastes.

David C. Little

is a professor at the University of Stirling. His research interests include in aquatic resource development.