Although this study focuses mainly on the fate of fish that escape from trawl codends during the capture process, it is useful for an overall understanding of the factors causing capture-induced stress, injury and mortality to take a brief look at studies conducted on the discard mortality of trawl-caught fish. Many of the factors that affect discarded fish are similar to those affecting escapees, but discarded fish have experienced additional stress and injury resulting from lifting on to the vessel deck, on-deck handling and air exposure, as well as the eventual discarding process.
It is well demonstrated that discarding is a widely applied practice in commercial fisheries and that there are many reasons for discarding (reviewed by FAO, 1983; 1994; 1997; 2004; Hall, 1996; Alverson, 1998; Hall, Alverson and Metuzals, 2000). It is also clear that not all of the discarded fish and other organisms survive (e.g. ICES, 2000; Davis, 2002). In fact, discard mortality represents a large source of uncertainty in estimates of overall fish mortality rates, and is a potentially important issue in fisheries management worldwide.
Most of the work conducted on discard mortality has concentrated on the survival of fish caught by trawls (otter trawls, beam trawls, shrimp trawls). A few studies include other gears such as longlines, gillnets and trap-nets (see page 25). Most discard mortality studies conducted in the field have focused on capture- and handling-induced stressors. Davis (2002) pointed out the important role of environmental factors, size- and species-related sensitivity to stressors, and interactions among stressors, all of which increase the mortality of discards. This chapter addresses those factors that are important for a general understanding of stress, injury and mortality associated with trawl capture processes. The discussion includes not only fish, but also other animals (in particular invertebrates) affected.
The main factors affecting the stress, injury and mortality of discarded fish are related to capture stresses, fishing conditions and biological attributes (Davis, 2002). Capture stressors include such factors as net entrainment, crushing, wounding, sustained swimming until exhaustion, and changes in pressure. Fishing conditions include towing time and speed, light conditions, water and air temperature, anoxia, sea conditions, time on deck, and various handling procedures. Biological attributes include behaviour, size and species. In addition, there are factors such as seabird predation near the surface and fish, mammal and invertebrate predation in the whole water column.
Although a discarded fish may survive the discarding process and immediate predation, a rapid fall to the sea bed may result in death owing to the physiological effects of increased pressure. Moreover, the disruption of schooling behaviour and the attraction of predators by visual, olfactory and mechanical cues from injured fish may significantly increase mortality (Ryer, 2002; Ryer et al., 2004). Delayed mortality of discards may represent a significant source of unobserved mortality. Clearly, a large number of factors affect discarded fish; many of these factors are the same as those that affect the mortality of fish escaping from fishing gear, but there are also several additional factors involved in discard mortality.
The types and extent of injuries caused by discarding are species-specific (ICES, 2000; Davis, 2002). While some species are highly sensitive to capture and discarding, others are capable of surviving these traumas. Fish with gas bladders and other organs that inflate after capture because of pressure changes may easily become trapped near the surface after discarding, and may therefore suffer complete mortality. Hence, it is not surprising that bottom dwelling round fish such as cod, haddock and saithe generally do not survive the discarding process well (e.g. Thurow and Bohl, 1976; Hokenson and Ross, 1993; ICES, 2000; Davis, 2002). It is worth noting that these fish may have at least three types of reactions to decompression:
(a) the swim bladder may become overinflated but remain intact;
(b) the swim bladder may be ruptured, but gas is kept in the abdominal cavity; or
(c) the swim bladder may rupture and gas be released through the ruptured abdominal wall.
In the last case, a fish that dives downwards may have a better chance of survival. Nevertheless, regardless of the level of damage to gas-filled organs, it is clear that the most effective way of reducing the mortality of such species would be to use selection measures that allow them to escape before being lifted to or near the water surface.
In some cases, these fish might benefit from the removal of excess gas from their swim bladders or abdominal cavities (e.g. Lee, 1992), but such measures are unlikely to be applicable in commercial situations. It is apparent that fishing conducted in shallow waters may result in somewhat lower discard mortality than deeper water fishing. In fact, Jurvelius et al. (2000) observed that mid-water trawling was far more lethal to trawl-caught and released pike-perch than trawling near the surface. It is also apparent that decompression speed affects the survival of discarded fish.
For fish that do not have gas bladders (e.g. sablefish, lingcod, flatfishes) mortality after release is more variable (e.g. FAO, 1994; Erickson et al., 1997; Erickson and Pikitch, 1999; Davis, 2002). Several species of flatfish, for example, appear to have relatively good chances of survival (e.g. Kelle, 1976; 1977; Neilson, Waiwood and Smith, 1989; Van Beek, Van Leeuwen and Rijnsdorp, 1989; Robinson, Carr and Harris, 1993; Pikitch et al., 1996), although relatively high mortalities have also been observed (e.g. Robinson, Carr and Harris, 1993; Lindeboom and de Groot, 1998). It is obvious that improved deck handling measures can be used to reduce further the discard mortality of these species.
The discard mortality of fish is related to the fragility and physical characteristics of each species. For instance, some fish species lose scales easily and may therefore suffer substantially higher mortality than species with more robust skin structures. If the catch consists of a mixture of species with hard parts, spines, shells and carapaces, and other more fragile species, the mortality of these latter species may be substantially greater than when only soft-bodied individuals are present. It is also notable that flatfish (e.g. Pacific halibut) may be more sensitive than round fish (such as walleye pollack, sablefish and lingcod) to suffocation in nets from pressure on the operculum (Davis, 2002). In addition, fish size may affect discard mortality. Smaller fish are generally weaker and more sensitive to capture and handling stress.
The average survival rate of undersized plaice, sole and dab in the German shrimp trawling was substantially lower than that of larger individuals (Kelle, 1976). The body size of trawl-caught and discarded Pacific halibut significantly affected mortality (Pikitch et al., 1996). The increased sensitivity of smaller fish may be caused by fatigue from the swimming forced by the gear and by abrasion and crushing injury. Smaller specimens may also be more sensitive to increased water temperature at the surface, as well as to all kinds of handling on board. These effects have not yet been investigated thoroughly enough to make any quantitative evaluations.
On-deck handling time and air temperature are among the most important factors potentially influencing the survival of discards. Exposure to air is almost unavoidable when the catch is brought on to the deck. Exposure times can range from a few minutes when catches are small to several hours when catches are large or handling is inefficient. Exposure may occur over a wide range of air temperatures.
Higher survival is associated with short air exposure times and low air temperature on deck (e.g. de Veen et al., 1975; Thurow and Bohl, 1976; Kelle, 1976; Pikitch et al., 1996; Nielson, Waiwood and Smith, 1989; Erickson and Pikitch, 1999). Direct sunlight on deck may markedly increase the mortality of discarded fish (e.g. Kelle, 1976; Pikitch et al., 1996). Any changes in fishing practices that reduce handling time and exposure to air would reduce discard mortality (assuming that lethal stress level has not been encountered).
It should be noted that freezing temperatures on deck may also contribute to discard mortality. However, very few scientific studies exist on this subject. Although handling time and air exposure are undoubtedly very important factors, it is obvious that the general handling process of discarded fish also plays an important role in determining their fate. It is also likely that sea state affects discard mortality. For instance, the haul back of trawl gear and the handling of catch on deck may take substantially longer in rough sea conditions, thereby increasing the likelihood of injury and mortality.
Water temperature affects the survival of discarded fish. There are marked species-specific differences in temperature sensitivity. Erickson et al. (1997) observed very high mortality (> 95 percent) for trawl-caught and discarded sablefish when surface water temperatures were high (18 to 20 °C). Mortality was substantially lower when surface water temperatures were 12 to 15 °C (Erickson and Pikitch, 1999).
Laboratory studies have shown similar results. Exposure of sablefish held at 5 °C to a range of seawater temperatures between 12 and 16 °C in the laboratory resulted in loss of feeding and increased physiological stress and mortality (Davis, Olla and Schreck, 2001). Mortality occurred after towing fish in a net at 5 °C and then exposing them to 12 °C. Exposure to 16 °C resulted in complete mortality for sablefish. Similarly, mortality rates increased with increasing seawater temperature for Pacific halibut and lingcod that were towed at 5 and 8 °C and then exposed to increased water temperatures, with 100 percent mortality reached at 18 °C in Pacific halibut and at 20 °C in lingcod (Davis and Olla, 2001; 2002).
Hyvärinen, Heinimaa and Rita (2004) exposed brown trout to an abrupt cold shock (10 min exposure to ice water in a chilling tank) after forced swimming that simulated swimming in a trawl. This treatment caused temporary unconsciousness of the fish. The cold shock seemed to be a stress factor, but the results did not indicate any permanently damaging effects on the fish. However, fish discarded in an unconscious state would certainly be highly vulnerable to predation by birds and predatory fish. Moreover, a longer exposure to cold water would likely be substantially more damaging.
Pacific halibut are frequently caught as bycatch by demersal trawlers in the Alaskan groundfish fishery (Figure 5). This bycatch must be released (discarded) back to the sea. Pikitch et al. (1996) estimated the mortality and physiological condition of trawl-caught and discarded Pacific halibut in the Gulf of Alaska.
Tow duration (of one to three hours) was among the factors that significantly affected halibut mortality. Similarly, the mortality of trawl-caught and discarded sablefish increased with longer tow durations (Erickson and Pikitch, 1999). These studies suggest that tow duration can be a significant factor affecting the survival of discarded fish. On the other hand, it is generally not known how long discarded fish have spent in the trawl. The duration of towing did not have a significant effect on the stress and mortality of undersized brown trout released after the haul (Turunen, Käkelä and Hyvärinen, 1994). Catch abundance, on the other hand, was a significant factor in causing stress.
Clearly, the effect of tow duration might be connected to codend catch size and catch composition, as well as to towing speed. It should be noted that catch volumes are rarely very large in experimental tows, so this variable is likely to come out as non-significant. Both longer tow duration and larger catch quantities increased the mortality of undersized discarded plaice (Kelle, 1976).
The effect of towing time on discard mortality is complex and should be explored more carefully. Nevertheless, for some species, shorter tows would probably increase the chances of survival. It is also likely that catch composition has an effect on discard mortality. Higher mortality may be assumed when the catch consists of spiny species and rubbish/trash. The amount of sand mixed in with the catch may also affect survival (Pikitch et al., 1996).
Hill and Wassenberg (1990) studied the survival of fish and invertebrates discarded in a prawn trawl fishery operating in Torres Strait, Australia. They found that none of the crustacean floated, and about half of them survived. All crabs survived the 12-hour monitoring period. Twenty-six percent of cephalopods floated, with a survival rate of only 2 percent. The majority (88 percent) of starfish survived the 12-hour monitoring period.
Wassenberg and Hill (1993) concluded that the major factors determining the fate of discards from shrimp trawlers are whether or not the animals were alive when discarded and whether they sank or floated. It is noteworthy that these authors later suggested that their previous assessment of survival rates for animals discarded from prawn trawlers may have been optimistic because they monitored for only 12 hours. They argued that longer experiments would have led to substantially lower survival figures, particularly for crustaceans. In many studies the estimated discard mortality of crustaceans has been high (e.g. Wileman et al., 1999) or very high (e.g. Harris and Ulmestrand, 2002). After being kept in tanks for two days, the average mortality for king and Tanner crabs captured as bycatch in commercial sole trawls in the Eastern Bering Sea was 78 to 79 percent (Steven, 1990). However, Lancaster and Frid (2002) estimated that about 80 percent of undersized brown shrimp survive the capture and discarding processes in the Solway shrimp fisheries (United Kingdom).
Harris and Ulmestrand (2004) demonstrated that nearly all discarded Norway lobsters in the Skagerrak-Kattegat Norway lobster fishery died because they were discarded through a low salinity surface layer. The authors underline the need to switch from discarding-based size-sorting lobster trawling methods towards ones that are size-selective. Norway lobsters escaping from a trawl codend have a relatively high likelihood of survival (Wileman et al., 1999).
In most cases, shelled molluscs and echinoderms appear to survive capture and discard processes relatively well (e.g. Hill and Wassenberg, 1990). The survival of trawl-caught Patagonian scallops subjected to 30 minutes of aerial exposure on board was also high (Bremec, Lasta and Hernandez, 2004). Sea urchins, on the other hand, may suffer high mortality in beam trawling (e.g. Kaiser and Spencer, 1995).
It is likely that the on-deck handling processes will affect the later survival of discarded invertebrates. For instance, the instant and delayed mortalities of trap-caught snow crab increased substantially with increased drop height on deck (Grant, 2003). Climatic conditions may have a strong impact. Smith and Howell (1987) assessed the mortality of trawl-caught American lobsters that were exposed to sub-freezing (-9.5 °C) temperatures for periods of 30 to 120 minutes. The mortality of lobsters reached 100 percent at 120 minutes exposure.
Generally, fish do not survive discarding processes well. However, the mortality of discards varies considerably according to species, size and environmental conditions, as well as with fishing and handling procedures. Air exposure on deck clearly affects stress and mortality. Higher survival rates are associated with short air exposure and low air temperature on deck. Exposure times are a function of handling times on deck, and can range from a few minutes to several hours, occurring over a wide range of air temperatures.
A number of other factors have also been shown to affect the mortality of discarded fish. Extreme sea surface temperatures were shown to have a significant impact on the mortality of discarded fish and invertebrates; hence, when possible, fishing seasons should be established to exclude extreme water (and air) temperatures. Smaller individuals are generally more sensitive to capture and discard-induced stressors and may therefore show greater discard mortality than larger specimens.
For species that do not have gas bladders that inflate after capture, discard mortality can be reduced through improved on-deck handling procedures and other operational improvements. Any changes in fishing practices that reduce handling time and exposure to air are likely to reduce mortality. For many species, however, it is difficult to obtain a significant reduction of mortality through improved handling processes.
To improve the survival of these fish, they should escape before they are landed on the vessel deck, preferably at the depth of capture. This would significantly increase their likelihood of survival. The use of selective fishing gears is a potential approach to reducing the mortality of these fish.