Resource removal – direct effects

The direct effects of the collection of wild “seed” for capture-based aquaculture are reviewed in this section of the report.

Overfishing

The main reason why species are chosen for capture-based aquaculture is their high market value. As a result of their value, most of these resources are already heavily exploited by commercial fisheries. Since the basis of capture-based aquaculture is the collection of “seed” from the wild stocks, this activity may increase the fishing effort on the target species. A particular danger arises for populations that are economically valuable but have low reproductive capacities because they mature at a large size.
FAO assessments of the various world fish stocks classify them in a range of categories from “under-exploited” to “overfished”. According to Hall (1999), overfishing can be divided into two types, recruitment overfishing and growth overfishing.
Recruitment overfishing occurs when a stock is depleted to a level where there is an unacceptable risk that the remaining adults will be insufficient to produce enough offspring to maintain the stock. This situation is most likely to occur in pelagic species where the individuals often form dense aggregations that can be easily detected, so that catches and catch rates can remain high even when the stock is severely depleted (Hall 1999). Additionally, many pelagic species are prone to dramatic natural fluctuations in recruitment success (e.g. the “anchoveta” in Peru and the herring fisheries in the North Atlantic, which both occur without warning).

The other type of overfishing is termed “growth overfishing”, which describes a state where fish are harvested at the wrong time in their life cycle. The extremes are the removal of a few larger older fish, or the capture of many small young fish. In between there is an optimal age at which the product of numbers and body size is maximal (Hall 1999).
Despite apparently substantial efforts to manage fisheries worldwide, there has been an almost universal failure to prevent the decline of fish stocks and the deterioration of the marine environment. Between 73-75% of the fish stocks globally offer no possibilities for increasing catches (FAO 2000).
The difficulties facing fisheries management in reducing fishing effort before the commercial extinction of the target stocks occur are immense. The characteristics of the life history of each species determines the level of fishing effort that will risk the survival of an exploited stock. Those characterized by short-life spans, rapid population growth and high reproductive output (R-selected species) respond rapidly to fishing and can cope with relatively high levels of mortality at young ages. Conversely, species with low natural mortality that allocate more energy to individual growth through competitive fitness than to reproduction (K-selected species), will support relatively low rates of fishing mortality and at older ages (Goni 1998).
Accurate assessment of the effects of overexploitation on a target population is not a simple task. In most cases this is due to the difficulty in separating natural and fishing-related mortality, and to the lack of stock assessment studies prior to the onset of exploitation. Where a direct link between stock collapse and over-exploitation has been established, natural changes (such as unusual hydrographical conditions) have also been seen to exist (Goni 1998; Masuda and Tsukamoto 1999).
The selected species that are considered in this report exhibit late reproduction, large size at reproduction, long-life spans, and form large spawning aggregations. This makes them vulnerable to overexploitation. In fact, the impact of intensive fishing is exacerbated by the K-selected life strategies of these genera and their tendencies to form predictable spawning aggregations. This may be critically important for population maintenance and the genetic diversity of the breeding stocks.
Heavy impacts on spawning aggregations are generally undesirable, and every attempt should be made to protect these brief, but important phases of the life cycle of these species from excessive disruption or exploitation. It is also essential to know how long the aggregations last, whether fish spawn throughout the entire period, and whether the same fish return repeatedly to the same site. Additional knowledge concerning the distance individuals travel to each aggregation site, and the proportion of any particular population involved in each aggregation, would facilitate management.
At the present time, none of the life cycles of the selected species is completely understood, and the biology of the species could cause more difficulties for stock evaluation. For an example, the removal of fingerlings from heavily fished adult populations may be an important factor contributing to the population decline of species such as the Hong Kong grouper E. akaara (Morris, Roberts and Hawkins 2000). This is likely to be significant because they spawn in limited areas throughout their geographic range. Sometimes this may simply reflect an area that is heavily exploited in general, but the possibility cannot be ruled out that some populations may be partially or fully self-recruiting and depend entirely on one or several aggregations (Rhodes and Sadovy 2002). Capture-based aquaculture seems to influence the status of some local grouper populations due to the “seed” collection for aquaculture practices. According to Sadovy (2000), capture-based grouper “seed” availability has declined in many areas of SE Asia, which may be in part be attributable to overfishing.
Carangid fish such as the greater amberjack have been heavily exploited because they form schools as an ecologically anti-predatory behaviour. Since 1993 (Andaloro 1993; Mazzola et al. 1993), some published fishery statistical data on the greater amberjack (S. dumerili) have indicated an over-exploitation of the juvenile classes in some areas of the South Mediterranean Sea. The availability of greater amberjack juveniles for capture-based aquaculture today comprizes a bottleneck for the development of this activity in Mediterranean countries. On the other hand, the status of the greater amberjack in the Gulf of Mexico has been estimated to be not overfished (NMFS 1998). Defining a species as overfished is difficult when several factors occur at the same time; this is the case for the Japanese amberjack (S. quinqueradiata) in its Pacific sub-population. Terauchi et al. (1991) observed a declining trend in adult fish stocks. This was probably due to the collection of larvae (“mojako”), which affected recruitment to the adult stock, coupled with a short-term decline due to an environmental factor (an abnormally low water temperature in the Pacific coastal sea area of Japan observed that year).
Most tuna stocks in temperate or tropical waters are under heavy pressure and are intensively or fully exploited. Some stocks are already overfished. Biological overfishing has been avoided on many stocks because of economic constraints and by transferring excess fishing capacity to other areas and oceans (South Pacific, Indian Ocean). By fishing further offshore on domes and thermocline fronts, the potential for increasing the exploitable biomass has reduced effort on more easily accessible, but less prolific stocks. After declines in the populations of bluefin tunas (northern and southern populations) were recorded, these stocks have been managed by regional bodies. The Convention for the Conservation of Southern Bluefin Tuna (CCSBT) was negotiated in 1994 in response to dramatic population declines. In the past, massive overfishing probably reduced the ability of the species to naturally replace itself and maintain healthy population levels (Buck 1995), so that today it is still considered an overfished resource. ICCAT (the International Commission for the Conservation of Atlantic Tunas) has defined two management units, West Atlantic tuna and East Atlantic tuna populations. Tudela (2002b) states that the western stock is overexploited and notes that the assessment of the East Atlantic bluefin tuna stock by ICCAT published in 1998 indicated that there had been a strong decline in the spawning stock biomass since 1993, as well as an increase in fishing mortality rates. The spawning biomass was estimated to be less than 20% of the 1970 level, and projections predicted a high probability of collapse within the following few years. The intense fishing pressure on small tunas seems to be contributing to overfishing and is reducing the potential long-term yield from the resource (Tudela 2002a,b,c). Today it is still difficult to evaluate the stock owing to lack of scientific data. It has been shown that it is difficult to detect overfishing or stock depletion risks in bluefin tuna, as spawning stocks and yields display conspicuous long term fluctuations. This is the result of a combination of year-to-year variations in recruitment and a long life span, as Fromentin and Fonteneau (2001) have shown using a mathematical model.
There has been a general reduction of the catches of glass eels of the European eel (A. anguilla) but recent studies have show that there is no actual decline in the total fishery yields along the Swedish west coast (Svedang 1999). Globally, the annual catch of glass eels of all species has gradually decreased over the past 25 years (Tanaka 2001) and a shortage of “seed” fish has become a very serious problem for eel capture-based aquaculture.

Recruitment success

Sensible exploitation of a fish stock requires management through legal and social instruments that in some way limit access to the resource. The fundamental biological aim for managing fisheries on a sustainable basis is that the catch rates should be balanced by recruitment. The problem is that for most stocks, recruitment cannot be simply predicted. While biological objectives have been the focus for fisheries biologists (and the sustainability of stocks is clearly a primary consideration), economic and social aspects of fisheries management also have profound effects on the choice of management regime, and the rigour with which it is imposed (Hall 1999).
For the species used in capture-based aquaculture, the problem of predicting recruitment is by far the most difficult facing fisheries biologists, who cannot be held solely responsible for the failure to manage fisheries successfully. Sound scientific advice is often not implemented because political and economic interests overturn it.
In pelagic spawning fishes such as groupers, where eggs are released into the water column to drift within surface currents, early natural mortality rates must be extremely high between egg production and settlement (when young fish change from their planktonic to their benthic phase) (Sadovy and Pet 1998). Estimates suggest that although each female grouper is capable of producing millions of eggs, only two young from each spawn will survive to adulthood under stable population conditions. What is not known is where the bulk of this early natural mortality occurs, and what the causes of this mortality are. If natural mortality remains high for some time after settlement, then the removal of young juveniles for capture-based aquaculture may have little impact on adult stocks, because most juveniles taken would otherwise perish due to natural causes. However, if early natural mortality rates have dropped to low levels prior to juvenile capture, then fishing mortality will represent an important source of total mortality (which is the sum of fishing mortality plus natural mortality).
Natural mortality drops rapidly during the early post-settlement period in tropical reef fish, i.e. several weeks or months following settlement. This strongly suggests that post-settlement mortality drops within a few weeks or months after settlement on a reef across a wide range of species and, moreover, that any harvest after this early period can negatively influence subsequent stock size (Sadovy and Pet 1998). Specimens taken for culture may be up to one year old at capture, and therefore many are probably caught well beyond the early weeks or months post-settlement. If this is the case, then fishing mortality represents a substantial proportion of total mortality and the fishery should be managed to avoid overfishing.
Fishermen (Figure 128) and researchers in the region agree that they see postlarvae in much greater numbers than grouper fry or fingerlings. This suggests that, as with the reef fish discussed above, there is considerable natural mortality among the postlarvae. Harvesting of postlarvae would thus have a lower impact on future adult populations than the harvesting of fry or fingerlings (Johannes and Ogburn 1999). However, the on-growing of a species in net cages intensifies fry collection in many areas and tends to reduce recruitment (Ahmad 1998).

Bycatch and discards

Most of the collection of “seed” material for capture-based aquaculture is carried out with traditional fishing gear. The aim for capture-based aquaculture is obviously to collect live fish for on-growing; the fishing technique selected should therefore be selective for the species and size of “seed” required for the aquaculture system. However, no gear is known to be one hundred percent selective for a given species or size range of individuals. Most gear and methods have some selectivity; their ability to select targets can be altered through modifications to design and operation. The catch in many fisheries thus consists of a mixture of target and non-target species. What does or does not comprize targets depends to a large extent on the market, and whether there are regulations in place prohibiting the capture of certain species or sizes of target organisms. Non-target species are often referred to as bycatch, a concept which is defined differently by numerous scientific bodies.
Hall (1996) defined bycatch as “that part of the capture that is discarded at sea, dead (or injured to an extent that death is the result)”. The word “capture”, in turn, means all that is taken in the gear. The capture can be divided into three components: 1) the “catch”, which is the portion retained for its economic value; 2) the “bycatch”, which is the portion discarded at sea already dead, and 3) the “release”, the portion released alive (Hall et al. 2000). The main reasons for discarding fishes (dead or alive), are as follows: the fish caught are the wrong species, size or sex, or are damaged; the fish are incompatible with the rest of the catch (from the point of view of storage); the fish are poisonous or spoil rapidly; there is a lack of space on board; “high grading”1 ; quotas have been reached; or the catch was of a prohibited species, in a prohibited season or fishing ground, or achieved with prohibited gear. Unfortunately, some of the gears used in the capture of species for capture-based aquaculture species can cause an incidental catch of non-target species (bycatch) and can collect undesirable sizes of target species.
1 Special cases of bycatch that are “high-grading” exist: these comprize the discard of a marketable species in order to retain the same species at a larger size and price, or to retain another species of higher value.

Figure 128. Fishermen repairing their fishing gear in the Philippines (Source: FAO)
The amount of the bycatch depends on the area, the period (season), and on the selectivity of the fishing gear. The bycatch issue is important for capture-based aquaculture species as it is one of the most significant of those affecting fishery management today. Different fishing techniques can lead to distinct types and rates of bycatch such as juvenile fish, benthic animals, marine mammals, marine birds, and vulnerable or endangered species, etc., that are often discarded dead. While bycatch and discard problems are usually measured as the potential loss of human food, the increased risk to a particularly vulnerable or endangered species (e.g. small cetaceans, turtles) is also significant. Bycatch can also affect biodiversity throughout impacts on top predators. For economists, its existence generates additional costs without affecting revenues, and may hinder long-term profitability. For fishermen, it causes conflicts among fisheries, gives them a bad public image, generates regulations and limitations on the use of resources, and has negative effects on the resources harvested through the mortality of juveniles and undersized individuals of the target species before they reach the optimum size. This problem must be addressed by scientists, fishery managers and members of the fishery industry. Although only a few fisheries include bycatches of the target species in their stock assessment, bycatch management will be an integral part of most future ecosystem management schemes (Hall et al. 2000). The total global discard (considering all the fisheries) is difficult to estimate. One assessment of the level of discards gave an estimate of 27 million tonnes in 1995 (FAO 1997c).
Besides the bycatches of fish (Figure 129), other animals may incidentally be captured with fishing gears (e.g. various species of whales, turtles and seabirds); although the level of the bycatches of such organisms seldom constitutes a threat to their population size, public concern makes it necessary to reduce them.
The level of bycatch associated with the collection of wild “seed” for capture-based aquaculture is not well documented. The same gear could cause different bycatch impacts depending on the area in which it is used. For example, catching bluefin tuna for capture-based aquaculture with the purse seining system in the Mediterranean is very efficient and does not entail high bycatches of cetaceans. This is not the case in other regions, such as purse seining in the Eastern Pacific. The best known example is the tuna-dolphin problem: incidental mortality of dolphins in tuna purse-seine fisheries in the Pacific Ocean during the 1960s was the first bycatch problem that received public attention. After the Marine Mammal Protection Act (MMPA) was introduced in 1972 (Gosliner 1999), dolphin mortality decreased from 133 000 in 1986 to only 1 877 in 1998 (Hall, Alverson and Metuzals 2000).
Improved selectivity can be achieved by modifying the gear design and/or operation, and by using alternative fishing gears. The capture of dolphins in the purse seine fishery for tuna has been reduced to an insignificant level by using a combination of technical changes, rescue techniques, the education of fishermen, and management actions. Experimental research is still going on in order to understand the potential danger represented by the “pinger”, an acoustic device that may disturb the dolphins.
Some of the capture-based aquaculture species are collected using floating objects or Fish Aggregating Devices (FADs); pelagic fish are often found in association with FADs as well as other animals (mammals, fish). Other natural structures (underwater mountains, etc.), artificial structures (wrecks or artificial reefs), or specially constructed FADs (like those used in Mauritius for game fishing or the “cheema” used in the Maltese “lampuka” fishery) are also effective. The reasons for this behaviour are still poorly understood, yet it is believed that by providing a substratum, smaller “feed” fish are initially attracted, which in turn attract the larger commercially valuable species.
Yellowtails are known to associate with FADs, and this is especially true for the greater amberjack (Seriola dumerili) and the Japanese amberjack (Seriola quinqueradiata). Since the 1980s tuna fishermen have been constructing and deploying artificial FADs, sometimes fitted with transmitter beacons to aid location. These electronically equipped FADs can the be deployed using new spatial strategies (Hallier 1995), and some also have echo-sounders that transmit information about the aggregated biomass by radio (Josse et al. 1999).
Harvesting fish associated with floating objects might threaten the pelagic ecosystem, due to various negative effects, such as an increased catch rate of juveniles or pre-reproductive animals, or an excessive mortality of non-targeted species (Hall 1998). A better understanding of these associations is therefore required to design and implement appropriate sustainable management procedures (Freon and Dagorn 2000).
Other gears and various fishing methods used for catching of “seed” for capture-based aquaculture operations (e.g. grouper) result in a high level of bycatch. For example, research carried out in Indonesia demonstrated that a very high percentage of the total catch captured in artificial reefs (called “gangos” in the Philippines) were non-target species, and that this method of harvesting can lead to a high bycatch mortality if not carefully handled (Mous et al. 1999). For many other gears used for grouper collection (e.g. fyke nets, scissor nets), bycatches during certain periods can be high. The bycatch comprizes a variety of fish sizes and species that are often thrown back at sea. The exception is in the densely populated areas of many developing countries, where the bycatch has a commercial value and is largely used for local consumption. In SE Asia this has serious implications, and the impact of “seed” fish for on growing on local foodfish resources cannot be ignored. For example, the bycatch of small juvenile rabbit-fish (Siganus spp.) is often high and represents a double loss, because in the same area the larger sizes of this species constitute a favoured food fish (Sadovy 2000).

Figure 129. Bycatch being delivered for fish feed in Thailand (Photo: FIIU-FI-FAO)
The use of trawls for eel fishing leads to a substantial bycatch. Due to their small mesh size, the trawl net affects many juvenile fish and up to 99% of the catch consists of species different from the target species, the eels (Hahlbeck 1994). The fyke net is another catch method that captures non-target species (Naismith and Knight 1994).

Direct physical disturbance and habitat destruction

Capture fisheries not only reduce the abundance of targeted stocks, but can have significant effects on the overall ecosystem and food chain, with consequences in other ecological and fishery-dependent systems, including those of mammals (Dayton et al. 1995). In addition, many nearshore ecosystems are substantially altered through habitat destruction caused by particular fishing methods.
The use of sodium cyanide, widely employed in the Philippines to catch groupers for capture based aquaculture, is contributing to the destruction of coral reefs (Goni 1998). This method not only causes direct damage to the habitat but also has collateral effects, including the death of non-target species of fish and invertebrates (Mous et al. 2000), as well as poor quality “seed”. The effects of poison on fish can be expected to be rather non-specific and alterations in the fish community structure and ecosystem appear likely.
Other gears used in Southeast Asia for the collection of “seed” for capture-based farmed species can be detrimental to near-shore habitats which are important nursery areas for many species. The use of scoop nets can cause significant impacts on the seabed and on benthic communities (Sadovy 2000). Benthic organisms are crushed, buried, or exposed to predators, and clouds of sediments arise. Alterations to the seabed biogeochemistry are also possible. Development management strategies that are designed to protect habitats are now established, and the use of scoop nets in several regions has now been banned or is regulated to reduce potential impacts.

Several organizations (e.g. the International Marine Life Alliance and the Nature Conservancy) have alerted coastal communities to the threat posed by destructive fishing, and “sustainable mariculture practices” and “best management practices” are rewarded at all points along the supply chain with increased prices and better market acceptance of products (Sadovy 2000; Hair, Bell and Doherty 2002).