ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on March 19, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(3):464-472; doi:10.1093/icesjms/fsm021
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Injury in trapped Dungeness crabs (Cancer magister)
1 Department of Biological Sciences, University of Rhode Island, 100 Flagg Road, Kingston, RI 02881, USA
2 Glacier Bay National Park, PO Box 140, Gustavus, AK 99826, USA
Correspondence to J. S. Barber: tel: +1 508 990 2860 ext. 148; fax: +1 508 990 0449; e-mail: jsbdives{at}hotmail.com
Barber J. S., and Cobb, J. S. 2007. Injury in trapped Dungeness crabs (Cancer magister). – ICES Journal of Marine Science, 64: 464–472.Although traps are the most effective fishing equipment used to capture crabs they can also result in indirect damage to target species. We examined the effect of trap-soak time, crab density, and the legal to sublegal size ratio on injury rates to male Dungeness crabs, Cancer magister. Our field results show that injuries increase significantly with increased trap-soak time, and as a consequence of different size ratios (crabs in traps with a greater ratio of sublegal crabs had more injuries). The injury rate was independent of density. In a laboratory experiment, injured crabs were as capable as intact crabs of obtaining, defending, and consuming food. However, studies on other crab species indicate that injury reduces growth, delays reproduction, decreases mating success, and increases mortality. If the costs of injury are similar for Dungeness crabs, this could diminish the rate of recruitment into the fishery.
Keywords: behaviour, Cancer magister, Dungeness crab, injury, trap fisheries, trap-soak time
Received 27 March 2006; accepted 21 December 2006; advance access publication 19 March 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Commercially exploited crustaceans are often captured in traps, a stationary type of gear that collects animals over time and holds them in a common space. The amount of time a trap is present in the water (the soak time) affects catch rate (Miller, 1979; Smith and Jamieson, 1989a). However, the effect of trap-soak time on the behaviour of targeted species remains largely unknown. As individuals continue to enter the trap, space and food could become increasingly limited and aggression may increase. Agonistic behaviour in decapod crustaceans can be an important factor influencing injury rates to conspecifics (Dingle, 1983; Hyatt, 1983). In addition to trap-soak time, other variables may affect the level of agonistic behaviour of trapped crustaceans, including sex and size distribution, density of conspecifics, and the availability of food (Miller, 1990).
Although some variability exists across species, aggressive interactions are generally won by larger individuals or by those possessing larger chelipeds than their opponent (Dingle, 1983). If this is the case, sublegal adult crustaceans unable to escape from traps may experience greater rates of agonistic encounter than larger, legal-sized individuals. The consequences of injury on the overall fitness of an individual are especially important when considering sublegal-sized animals, which will be released upon retrieval of the trap. Several studies have demonstrated that injury in decapod crustaceans can have broad-reaching consequences, including reduced growth rate (Davis, 1981), decreased reproductive success (Abello et al., 1994), increased risk of predation and cannibalism (Berzins and Caldwell, 1983), and decreased foraging ability (Juanes and Hartwick, 1990; Smith and Hines, 1991; Juanes and Smith, 1995). Although agonistic behaviour in decapod crustaceans has been widely researched (Dingle, 1983; Hyatt, 1983), little work has focused on the effects of aggression on and injury of crabs caught in traps. Moreover, appendage injuries have been documented in trapped crabs (Shirley and Shirley, 1988; Kimker, 1994), but the sources of these injuries have only been hypothesized.
The Dungeness crab (Cancer magister) supports trap fisheries throughout its range from California to Alaska (Kondzela and Shirley, 1993). The widespread distribution and the economic importance of this fishery led us to select the Dungeness crab as a model species for investigating the effects of trap-soak time on injury rate. Few studies have focused on the behaviour of Dungeness crabs within traps and the effect of trap-soak time on agonistic interaction, and very little research has investigated the consequences of injury on the behavioural ecology of Dungeness crabs.
To investigate factors influencing injury to Dungeness crabs and consequences for released injured males, three hypotheses were tested: (i) injury rates are related to trap-soak time and crab density; (ii) injury rates are related to trap-soak time and the ratio of legal to sublegal crabs in the trap; and (iii) injuries sustained in traps result in a decreased ability of sublegal males to compete for food.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Field experiments
To address the hypotheses that injury rates are related to soak time, we conducted two field experiments consisting of (i) soak time and density trials and (ii) soak time and legal: sublegal size ratio trials. All fieldwork was done in Bartlett Cove (58°30'N 135°53'W), a small estuarine embayment near the mouth of Glacier Bay, Alaska, USA. We collected the crabs for all our experiments using commercial Dungeness crab pots (0.91 m diameter, escape vents sealed). The carapace condition of the retrieved Dungeness crabs was classified as in Shirley and Shirley (1988). Male crabs were measured to the nearest millimetre directly anterior to the tenth antero-lateral spine (Shirley and Shirley, 1988), and the width and length of the left propodus was recorded.
We selected two size classes of male crab (sublegal, 152–165 mm carapace width [CW]; legal, 168–184 mm CW) for use in trials. The minimum legal size in the southeast Alaska Dungeness crab fishery is 165 mm CW. Sublegal crabs with a CW of 152–165 mm are likely to become trapped despite the presence of escape rings (Smith and Jamieson, 1989a). The legal crab size class was developed around the mean size of legal crabs we caught in Bartlett Cove (mean = 174 mm CW, s.e. = 0.13, n = 2131). Legal crabs and their chelae were significantly larger than the sublegal crabs used in the experiments (Table 1). Crabs with major injuries, including missing or regenerating appendages, missing eyes, cracks or breaks >5 mm long in the exoskeleton, or with a carapace condition classified as "very old" or "soft" were not used in the experiments. Crabs meeting the physical requirements were tagged with uniquely numbered, double-T Floy tags (Floy Tag and Manufacturing Company, Seattle, WA, USA). Using guidelines established by Smith and Jamieson (1989b), tags were placed through the posterior epimeral suture line such that the tags would be retained through ecdysis. Tagging crabs prevented their use in more than one experiment and facilitated the identification of individual crabs at the start and end of a trial.
|
At the start of a trial, crab injuries were assessed immediately prior to the placement of a crab in the experimental trap. We recorded injury location as specifically as possible, including the name of the appendage or area and the segment of the appendage. For all experiments, chelae were (i) bound closed with lobster bands, to remove the primary apparatus used by crabs in aggressive behaviour, or (ii) unbound. Bait (Pacific herring) was hung in the trap to mimic realistic trap conditions. We sealed all escape vents and entrances to prevent crab movement in and out of traps. Additionally, thick mesh netting was placed in the bottom of the traps to prevent crab legs from falling through the trap bottom while it was lowered to the benthos. Once the crabs were added, the traps were closed and lowered to the bottom, and left undisturbed for the designated soak time.
We classified new injuries into three categories: missing appendages, large injuries (>5 mm diameter), and small injuries (<5 mm diameter) (Figure 1). Large injuries usually involved damage to the shell so severe that internal tissues were visible. Smaller injuries generally consisted of circular puncture wounds or narrow cracks in the shell. Careful observations ensured that mating marks created by the action of a male carrying a female in a pre-mating embrace were not counted as injuries (Jamieson, 1996).
|
A procedural control experiment was conducted prior to soak-time experiments to test the assumption that dropping and lifting pots stocked with crabs would not affect the injury rate. Results demonstrated that injuries were not an artifact of trap setting, so the planned experiments were conducted (Barber, 2004).
Two main field experiments were carried out: (i) soak time and density and (ii) soak time and size ratio. We used 5- and 20-d soak times for both density and size-ratio experiments (Table 2). These soak times were selected to mimic the mean soak time of a commercial crabber in southeast Alaska (3–5 d soaks) and the mean soak time of a recreational crabber (10–30 d) (JSB, pers. obs.). The traps were set in a random order of treatments throughout the season from July through October 2002. To test the hypothesis that injuries increase with soak time and density, we experimented with high and low crab densities (20 or 5 crabs; Table 2). A 20-crab density is representative of the average catch of crabs in a trap that had been fishing for more than 2 d. The 5-crab density was chosen to represent a smaller catch of crabs (JSB, pers. obs.). Density trials used only sublegal crabs in order to standardize crab size in these experiments. The size-ratio trials held density constant at 12 crabs per trap, but modified the ratio of legal vs. sublegal crabs within the trap (Table 2). Three size ratios (legal: sublegal) were studied: 12:0, 8:4, and 4:8. Each size-ratio treatment was soaked for either 5 or 20 d.
|
Statistical analyses for field experiments
For all analyses, we considered each individual trap as a replicate, using the mean number of new injuries per crab per trap as the dependent variable. All traps were assumed to be independent, because they were separated from one another by at least 10 m in the field. We used SPSS 11.0 Statistical Software for all analyses.
First, we tested if soak time, crab density, and bound vs. unbound chelae influenced injury rates using a three-factor ANOVA with fixed factors. Using the same statistical test, we also investigated the effect of soak time, size ratios, and bound vs. unbound chelae on injury rates. Data were log10 (x + 0.1) transformed when necessary, and post hoc comparisons were conducted with a Tukey HSD for all significant main effects and interaction terms. Additionally, for each experiment we conducted four three-factor ANOVAs on the following response variables: all injuries combined, autotomized appendages, large injuries, and small injuries.
Prior to conducting the three-factor ANOVA on the soak-time and size-ratio experiments, we determined whether variation existed between the number of injuries received by legal vs. sublegal crabs in the experiments. Injuries could not be analysed on an individual basis because crabs within traps were not independent. Therefore, a single number was required for each trap replicate that represented the difference between sublegal crab injuries and legal crab injuries. This injury index was determined by subtracting the average number of legal crab injuries per pot (for all injuries combined) from the average number of sublegal crab injuries per pot (i.e. sublegal – legal = index). If the difference between the average number of injuries was close to zero, there would be no distinction between the number of injuries that legal crabs received vs. the number received by sublegal crabs. Given that no significant difference (between legal and sublegal crab injuries) was found using this index, we combined the original injury data from both legal and sublegal crabs for the three-factor ANOVA on the soak-time and size-ratio experiments (Barber, 2004).
Laboratory experiments
To determine the effect of injuries on a crab's ability to compete for resources, we conducted two laboratory-feeding trials. We obtained 62 sublegal and 24 legal crabs in late July from a commercial crabber. Sublegal and legal crabs meeting the same physical requirements as field-experiment crabs were tagged with a unique number. We recorded CW and cheliped dimension in addition to existing injuries as specified above in the guidelines for field experiments. Animals were maintained in tanks at the NOAA Auke Bay Laboratory in Juneau, Alaska, USA. All tanks were supplied with ambient running seawater and each crab had a separate compartment to exclude interactions with conspecifics. All experiments were conducted under fluorescent laboratory lighting during the day. For all trials, a running seawater tank was fitted with a large, sand-filled, plastic tub placed in the centre. Plastic walls were placed tightly around the tub to create a 35 x 80 cm trial area.
The first set of experiments examined the ability of injured crabs to compete for resources relative to the performance of intact crabs. As cheliped loss is the most common injury in C. magister (Shirley and Shirley, 1988), the competitive ability of crabs with two types of injuries was studied: loss of one cheliped or loss of both chelae. Three trial types were conducted with two crabs per trial: (i) intact vs. intact, (ii) one cheliped vs. intact, and (iii) no chelae vs. intact. All crabs used in these trials were in the sublegal size class (mean CW = 160 mm, s.e. = 0.31). No crabs were re-used in the experiment.
We conducted an additional experiment on the effect of size on competitive ability, in which intact legal-sized crabs were placed in competition with intact sublegal-sized crabs. Owing to limitations on laboratory space, all intact sublegal crabs used in this size experiment had been used previously in the intact vs. injured sublegal crab trials. However, the sublegal crabs had not been exposed to the legal-sized crabs prior to the experiments, were selected randomly, and had been subject to a minimum time lapse of seven days between use across experiment types. Additionally, both sublegal and legal crabs were only used once in this particular experiment. Considering the lack of complete independence among crabs across experiments, the results from the legal vs. sublegal crab experiment should be interpreted conservatively.
Two weeks prior to the beginning of the trials, 20 sublegal crabs were selected randomly, and injuries were induced by compressing the cheliped along the natural autotomy plane (Simonson and Hochberg, 1986). Crabs not selected for injury were handled for the same duration as the injured crabs, but no injuries were induced. All crabs were given, at a minimum, two weeks to recover from the stress of handling or injury.
We selected food as a model resource for these competition trials because of the ease of providing an equal quantity of food in each trial and because food is a resource that crabs are likely to compete for in the field. The crabs were randomly chosen for use in each trial, fed a week before their trial, and starved for six days in order to standardize feeding history. Then they were transferred to the experimental tank, where an opaque partition separated the two crabs for 15 min. After the acclimation period, the partition was removed and a slice of herring (mean wet weight 22 g) was placed an equal distance away from both crabs. The full duration of each trial (80 min) was recorded on digital video and a trained observer was always present. The crab that obtained >50% of the food during the trial was considered the "winner" of the resource. We determined if a crab ate >50% of the food through careful observation. With one exception, it was always obvious which crab possessed and ate most of the food. For the trial where it was difficult to determine which crab ate >50% of the food, we compared the total amount of time each crab possessed the food, observer notes, and video notes to determine the "winner".
Statistical analyses for laboratory experiments
Chi-squared analyses were conducted on the intact-sublegal crab vs. injured-sublegal crab trials and on the intact-legal vs. intact-sublegal trials. For the analyses, our hypothesis was that no difference would exist in the ability of any crab to obtain >50% of the food (i.e. they had a 50/50 chance of obtaining food). A separate 
2 test, also predicting a 50/50 outcome, was used to determine whether crabs that obtained the food first subsequently lost most of the food to their opponents. Because no physical difference existed between crabs in the intact-sublegal vs. intact-sublegal trials, they were not included in these 2 analyses. Data from these particular trials were included in the analyses described below.
We used a single-factor ANOVA with the time-to-food acquisition as the dependent factor and the injury state of the crab as the fixed factor. This allowed us to test for differences in the time it took a crab of a particular physical status to acquire resources over all trials.
To further explore other factors related to feeding and injury, we analysed the amount of time injured-sublegal crabs, intact-sublegal crabs, and intact-legal crabs spent holding and eating the food. Such analysis examined the data for differences in the total time spent with food (holding and eating combined) using a single-factor ANOVA. The intact-sublegal crab data from the legal vs. sublegal trials were excluded from all analyses because of the re-use of those crabs in the size experiments.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Field experiments
Soak time and crab density
When all injuries were combined for analysis, the mean number of new injuries per crab per trap was significantly different between unbound-cheliped (mean = 0.93, s.e. = 0.11, n = 19) and bound-cheliped treatments (mean = 0.156, s.e. = 0.03, n = 13; Table 3; Figure 2). As soak time increased, injuries increased (5-d mean = 0.38, s.e. = 0.07, n = 16; 20-d mean = 0.85, s.e. = 0.15, n = 16; Table 3; Figure 2). Injuries were density-independent in traps (5-density mean = 0.56, s.e. = 0.13, n = 17; 20-density mean = 0.67, s.e. = 0.14, n = 15; Table 3; Figure 2). There was a significant relationship between the cheliped-binding treatment and soak time (Table 3).
|
|
There were no significant differences between the number of autotomized appendages in any of the single factors or interactions (Table 3), but there were significantly fewer small injuries in the bound-cheliped treatment (mean = 0.09, s.e. = 0.03, n = 13) than in the unbound treatment (mean = 0.50, s.e. = 0.09, n = 19; Table 3; Figure 2). The extent of large injuries also differed between traps containing crabs with bound chelae (mean = 0.05, s.e. = 0.03, n = 13) and those containing crabs with unbound chelae (mean = 0.33, s.e. = 0.06, n = 19; Table 3; Figure 2). As soak time increased, the number of large injuries also increased (5-d mean = 0.12, s.e. = 0.03, n =16; 20-d mean = 0.31, s.e. = 0.08, n = 16; Table 3; Figure 2). A significant relationship was detected with large injuries between soak time and bound or unbound chelae (Table 3; Figure 2).
Soak time and crab size ratio
As expected, injuries were more frequent when crabs in traps had unbound chelae. When all injuries were combined, significantly more injuries were found in the unbound-cheliped treatments (mean = 0.89, s.e. = 0.07, n = 32) than in the bound-cheliped treatments (mean = 0.23, s.e. = 0.04, n = 21) (Table 4; Figure 3).
|
|
As with the crab-density experiments, soak time affected injury rate. More new injuries were detected in the longer soak time than the shorter soak time (5-d mean = 0.52, s.e. = 0.10, n = 27; 20-d mean = 0.74, s.e. = 0.09, n = 26) (Table 4; Figure 3). The number of autotomized appendages also increased significantly with greater soak time (5-d mean = 0.01, s.e. = 0.009, n = 27; 20-d mean = 0.05, s.e. = 0.012, n = 26; Table 4). However, there was no significant difference in the number of autotomized appendages between traps with unbound and bound chelae (Table 4).
Traps with more legal-sized crabs had fewer injuries. For all injuries combined, traps with 4 legal and 8 sublegal crabs (mean = 0.73, s.e. = 0.10, n = 17) had significantly more injuries than traps with only legal-sized crabs (mean = 0.51, s.e. = 0.10, n = 19) (Tukey HSD, p = 0.013; Figure 3). Traps with 8 legal and 4 sublegal crabs (mean = 0.66, s.e. = 0.12, n = 17) yielded similar results to those of the other treatments (Tukey HSD, p > 0.05 for both comparisons).
Small injuries were more frequent in unbound-cheliped traps (mean = 0.57, s.e. = 0.06, n = 32) than in bound-cheliped traps (mean = 0.13, s.e. = 0.03, n = 21; Table 4). Similarly, large injuries were more frequent in unbound-cheliped traps (mean = 0.28, s.e. = 0.04, n = 32) than in bound-cheliped traps (mean = 0.08, s.e. = 0.02, n = 21; Table 4).
Laboratory experiments
When in apparent competition for food in laboratory tanks, injured crabs performed similarly to intact crabs in their ability to obtain food during all treatments. Sublegal crabs with one cheliped (SOC) were equally capable of obtaining more than half the food when competing against a sublegal crab with both chelae (SBC) (SOC = 70%, SBC = 30%, 2 = 1.6, n = 10, p = 0.21; Figure 4a). Sublegal crabs with no chelae (SNC) also had similar ability to obtain most food in competition with intact sublegal crabs (SNC = 60%, SBC = 40%, 2 = 0.40, n = 10, p = 0.53; Figure 4b). Legal crabs with both chelae (LBC) also had similar ability to obtain food as intact sublegal crabs (LBC = 57%, SBC = 43%, 2 = 0.14, n = 7; p = 0.71; Figure 4c). Injured crabs or intact legal crabs did not initially obtain food more often than intact sublegal crabs (four separate 2 analyses, p > 0.05). Therefore, crabs that obtained food first did not subsequently lose most of the food to their opponents.
|
There was no difference in the amount of time required by injured crabs to find and acquire food compared with the time spent by intact crabs (F3, 32 = 0.546, p = 0.66; Figure 5). No significant differences were found between intact sublegals, intact legals, and injured sublegal crabs in terms of total time spent with food, i.e. holding and eating combined (F3, 61 = 0.14; p = 0.94; Figure 6).
|
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The field experiments demonstrated that duration of soak time influences the rate of injury in Dungeness crabs. This result was expected, given the agonistic nature of the species (Jacoby, 1983) and the fact that an earlier study had noted a positive relationship between injury and soak time (Shirley and Shirley, 1988). Understanding the reasons why injuries increase with soak time, however, requires closer examination of crab behaviour.
Increased competition for resources in the trap is not likely to explain the increase in injuries because (i) mating was not possible because there were only males in the traps, and (ii) there was no positive relationship between injuries and crab density. Competition for food seems a plausible explanation, although bait in the traps was consumed within the first few days of each experiment. Additionally, we only recorded the presence of bycatch in three out of 83 treatments, suggesting that opportunities for competing over other food resources were scarce. Despite this, competition over small fragments of food could account for some of the injuries. Further, it is unlikely that crabs attempted to cannibalize trapped conspecifics. If cannibalism had been taking place, there should have been greater mortality within the unbound-cheliped traps than in the bound-cheliped traps. The number of crabs that died within traps was equal regardless of treatment (50/50 expected, 2 = 0.6, n = 15; p = 0.439). Consequently, cannibalism, although believed to take place in traps (Shirley and Shirley, 1988), did not appear to be a factor regulating the outcome of injuries to the experimental crabs. An alternative explanation for the observed increase in injuries with soak time considers how behavioural changes may influence the level of aggression and the number of agonistic interactions within the trap through time.
Qualitative observations from a video camera mounted over a crab pot revealed that crab behaviour seemed to change with an increase in soak time (Barber, 2004). Although the turbidity of the water prevented us making planned quantifiable observations, we were still able to observe unique behavioural interactions. During a 20-d soak, crabs appeared to be quite active both in the presence of food at the beginning of the soak and towards the end of longer soak periods, while crabs seemed less active immediately after consuming the bait. These observations suggest that new injuries could be attributable to agonistic interactions over bait at the beginning of a soak, and that the increase in injuries in the longer soak-time experiments may be a reflection of the observed secondary escalation in movement and interactions towards the end of a longer soak.
Our results also demonstrate that injuries to trapped crabs are density-independent. It is plausible that increases in crab density could cause a shift in behaviour that minimizes injury to the population. Dunham (1972) showed that Homarus americanus exhibit decreased agonism when held in tanks at high density. Further, T. C. Shirley, S. J. Taggart, and J. Mondragon (unpublished data) recorded an inverse relationship between limb loss and catch rate in a long-term field study of Dungeness crab. This suggests that when crab density is high, the rate of agonistic interaction is low, or alternatively that most limb loss is related to activities other than agonistic interaction. In our qualitative observations with underwater video, there appeared to be fewer interactions and less movement in traps with higher crab densities than in traps with lower crab densities (Barber, 2004).
Larger decapod crustaceans tend to be aggressively dominant over smaller conspecifics (Dingle, 1983; Hyatt, 1983). Previous work on Dungeness crab reported that larger animals often won agonistic interactions with smaller ones (Jacoby, 1980), so we predicted that smaller crabs would sustain greater rates of injury in our experiments. However, size class did not affect injury rates in the field, and our laboratory results suggest that sublegal- and legal-size crabs had similar ability to compete for food.
On the other hand, more aggressive behaviour by smaller crabs could be creating some of the differences we observed. Traps stocked with a ratio of 8 sublegal to 4 legal crabs had significantly more injuries than traps with 12 legal-sized crabs. Smaller crabs (mostly juveniles) have been observed in more intense, highly physical, agonistic interactions when competing with a crab of similar size (Jacoby, 1980). Larger crabs, however, performed ritualized, less forceful interactions in trials with crabs of equal size (Jacoby, 1980). Previous field research has shown that sublegal males receive significantly more injuries than legal-sized males (T. C. Shirley, S. J. Taggart, and J. Mondragon, unpublished data), perhaps reflecting similar trends in behaviour in which sublegal crabs may be inflicting injuries upon other sublegal crabs during agonistic interaction. Therefore, ontogenetic changes in behaviour could explain why traps with more small crabs sustained greater rates of injury. This finding could have important management implications, because traps containing many crabs of sublegal size may also have some of the highest rates of injury.
The significant interactions found in the soak time and density trials are more difficult to interpret. Our result implies that soak time and cheliped treatments cannot be interpreted separately. We believe that the interaction may be attributable to other causal agents of injury, rather than agonistic behaviour by trapped crabs. Our video records showed free crabs approaching an open fishing pot and interacting with trapped crabs through the sides of the pot (Barber, 2004). Miller (1978) reported similar behaviour of free C. productus aggressively interacting with trapped conspecifics through the mesh wall of the trap. The new injuries we recorded in the bound-cheliped treatments usually consisted of open wounds to the exoskeleton, an injury easily induced by grasping chelae and not normally created by interaction with other predators. This behaviour may explain the recorded increase in injuries in the 20-crab bound-cheliped traps, where crabs were more crowded against the wall of the trap and exposed to interactions with untrapped crabs.
The results of the laboratory food-competition trials were surprising in that injured crabs seem equally capable of competing for food with intact crabs of similar size. The injured crabs appeared to use more intense action patterns during interactions than the intact crabs (see Jacoby, 1980, for descriptions of action patterns; JSB, pers. obs.), and the first walking legs were used in agonistic interactions as if they were chelae (JSB, pers. obs.). These modified behaviours probably resulted in sustaining an injured crab's competitive ability. Moreover, we found a trend suggestive of injured crabs obtaining food more often than intact animals (Figure 4a and b). This counterintuitive result requires further investigation with careful consideration of the costs and benefits associated with obtaining food.
Our laboratory results should stimulate future research to consider the costs of injury to Dungeness crabs. For example, given that injured crabs are equally capable of competing for and eating soft food as intact crabs, does this mean that the consequences of injury will be similarly unaffected in terms of growth rate or other life-history parameters? Further, it is apparent that one cannot assume that injured C. magister will behave in the same manner as other species of injured crab (see Juanes and Smith, 1995, for a review of injured crab behaviour). Future studies should investigate other possible ecological consequences of Dungeness crab behaviour with respect to limb damage.
Implications
A strong market for Dungeness crabs has led to intense commercial fishing effort, resulting in an increase in the handling of sublegal crabs (Breen, 1987; Miller, 1990). California is currently the only state to regulate the trap-soak time of commercial crabbers (CDFG, 2005). Therefore, two primary questions arise in response to the potential impact that longer trap soak times may have on these fisheries: (i) will autotomy or limb damage affect the life history and behavioural characteristics of Dungeness crabs, and (ii) are enough sublegal crabs caught in commercial and recreational fisheries to merit concern over injuries?
Although our study only considered the impact of cheliped autotomy on competition for food, autotomy clearly affects other life history, behavioural, and functional parameters. In a long-term field study of Dungeness crabs, many were recorded with autotomized appendages but very few with regenerating limbs, suggesting that survival is poor in crabs with autotomized appendages (T. C. Shirley, S. J. Taggart, and J. Mondragon, unpublished data). The impact of limb autotomy is likely to be much greater in decapod crustaceans with long intermoult periods, such as Dungeness crabs (Kondzela and Shirley, 1993), because it may take up to three years for a single limb to regenerate fully. Injured C. magister could also exhibit reduced growth, delayed reproduction, decreased mating success, and greater mortality. However, until these hypotheses are tested directly, it would be inappropriate to assume that the effects of autotomy on Dungeness crabs will be similar to the effects on other decapod crustaceans.
The most common new injuries to trapped crabs were large, open wounds to the exoskeleton, or missing appendage tips. Injured crustaceans may be more vulnerable to infection (Stewart, 1993), and diseased Dungeness crabs have been recorded with new wounds in their exoskeleton (Armstrong et al., 1981). Therefore, increased injury through greater soak time or from different size ratios may expose trapped Dungeness crabs to a greater risk of infection, because the crabs are unnaturally aggregated and infection could spread easily throughout a trap (Armstrong et al., 1981). In addition to disease, partial limb loss (i.e. the loss of a limb segment such as the dactyl) can result in the regeneration of a deformed segment (Juanes and Smith, 1995). Dungeness crabs search for bivalves by prodding their dactyl tips into the substratum to receive tactile stimuli, then recover the bivalve by digging with legs and chelae (Pearson et al., 1981), so damage or loss of appendage tips could impede a crab's foraging ability.
The sublegal crabs in our experiments ranged in size from 152 mm to 165 mm CW, representing the size of crabs most likely to be retained in traps despite the presence of escape rings (Smith and Jamieson, 1989a; Miller, 1990). Several studies have reported very different estimates of the ratio of sublegal crabs to legal-sized crabs caught in commercial fishing traps with escape rings (High, 1976; Breen, 1987, Smith and Jamieson, 1989a). Using Smith and Jamieson's (1989a) data, Miller (1990) estimated the ratio of sublegal to legal-sized crabs to be 2.6:1. The size ratios of crabs caught in traps will also vary within a fishing season, where the catch of legal crabs will decrease through time (Smith and Jamieson, 1989b; Rumble and Bishop, 2002), and the sublegal catch will presumably increase (Smith and Jamieson, 1989b). If fishers respond to catching a larger proportion of sublegal crabs by lengthening soak times, our results would suggest that an unintended consequence could be greater rates of injury to the crabs. Given that injury to sublegal crabs is common and could result in greater fishing mortality as well as the loss of potential future revenue, effort to reduce the negative impacts of trapping should focus on portions of the season when the sublegal catch is large.
There is also little information on the impact of recreational fishing on sublegal-sized crabs. It is quite common for recreational pots in Glacier Bay to be soaked for 3–4 weeks at a time, whereas commercial traps (near Glacier Bay) generally have soak times of 3–5 d (JSB, pers. obs.). Our study illustrates clearly that soak time can influence injury rates to local crab populations, so it is likely that recreational crabbers also increase the rate of injury to sublegal crabs. Although the Washington Department of Fish and Wildlife (WDFW, 2005) recently instituted mandatory catch record-keeping for recreational crabbers, most regulatory agencies do not keep track of the sublegal catch aside from annual surveys, simply requiring that all crabs below minimum legal size be returned to the water upon retrieval of the trap (Rumble and Bishop, 2002). Therefore, the catch of sublegal-sized crabs throughout the range of the fishery could be high enough to merit concern about trap-related injury and its potential effect on mortality rates.
|
Acknowledgements |
|---|
This project would not have been possible without help from field assistant Kathleen Lotterhos. We also acknowledge the support of B. Eichenlaub, M. Donnellan, J. Fisher, M. B. Moss, J. Smith, and C. Soiseth from Glacier Bay National Park, and G. Bishop and J. Rumble from the Alaska Department of Fish and Game. The NOAA Auke Bay Laboratory provided space and tanks. Capt. Allen Morin of the FV "Jenny" kindly donated his fishing time and collected crabs for laboratory experiments. We are especially grateful to A. Moles, R. Stone, and S. Wells for their assistance in the laboratory, and also acknowledge the support of A. Andrews, J. Bodkin, J. de la Bruere, G. Eckert, N. Hobbs, P. Hooge, G. Moeser, J. Mondragon, P. Paton, T. Shirley, J. Taggart, C. Comeau, and countless field volunteers. Our manuscript was greatly improved by the insightful comments of J. Dimond, J. Grabowski, R. J. Miller, and one anonymous reviewer. Support for the project was provided by the National Park Service and the PADI Project AWARE Foundation. The research was completed in partial fulfilment of a MS degree for JSB at the University of Rhode Island.
|
Footnotes |
|---|
Present address: Massachusetts Division of Marine Fisheries, 1213 Purchase Street, 3rd Floor, New Bedford, MA 02740, USA.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Abello P., Warman C. G., Reid D. G., Naylor E. (1994) Chela loss in the shore crab Carcinus maenas (Crustacea: Brachyrua) and its effect on mating success. Marine Biology 121:247–252.[CrossRef]
Armstrong D. A., Burreson E. M., Sparks A.K. (1981) A ciliate infection (Paranophrys sp.) in laboratory-held Dungeness crabs, Cancer magister. Journal of Invertebrate Pathology 37:201–209.[CrossRef][Web of Science]
Barber J. S. (2004) Factors influencing injury of trapped Dungeness crabs and survival consequences for released males. MS thesis,(Department of Biological Sciences, University of Rhode Island, Kingston, RI, USA).
Berzins I. K. and Caldwell R. L. (1983) The effect of injury on the agonistic behaviour of the stomatopod, Gonodactylus bredini (Manning). Marine Behaviour and Physiology 10:83–96.
Breen P. A. (1987) Mortality of Dungeness crabs caused by lost traps in the Fraser River Estuary, British Columbia. North American Journal of Fisheries Management 7:429–435.[CrossRef]
CDFG (California Department of Fish Game). (2005) California Fish and Game Code Section 9004. , Sacramento CA 95814.
Fishery Bulletin US Davis G. (1981) Effects of injuries on spiny lobster, Panulirus argus, and implications for fishery management. 78:979–984.
Dingle H. (1983) Strategies of agonistic behaviour in crustacea. In Rebach S. and Dunham D. W. (Eds.). Studies in Adaptation(John Wiley, New York) pp. 85–111 300.
Dunham P. J. (1972) Some effects of group housing upon the aggressive behaviour of the lobster Homarus americanus. Journal of the Fisheries Research Board of Canada 29:598–601.[Web of Science]
High W. L. (1976) Escape of Dungeness crabs from pots. Marine Fisheries Review 38:19–23.
Hyatt G. W. (1983) Qualitative and quantitative dimensions of crustacean aggression. In Rebach S. and Dunham D. W. (Eds.). Studies in Adaptation(John Wiley, New York) pp. 113–139 300.
Jacoby C. A. (1980) Ontogeny of behaviour in the Dungeness crab, Cancer magister Dana 1852. PhD thesis,(Stanford University, Hopkins Marine Station, CA).
Jacoby C. A. (1983) Ontogeny of behaviour in crab instars of the Dungeness crab, Cancer magister Dana 1852. Journal of Comparative Ethology 63:1–16.
Jamieson G. S. (1996) Molting patterns in southern British Columbia Dungeness crab and implications for fisheries. High Latitude Crabs: Biology, Management, and Economics (University of Alaska, Fairbanks)397–410 Alaska Sea Grant College Program Report, 96–02 713.
Juanes F. and Hartwick E. B. (1990) Prey size selection in Dungeness crabs: the effect of claw damage. Ecology 71:744–758.[CrossRef][Web of Science]
Juanes F. and Smith L. D. (1995) The ecological consequences of limb damage and loss in decapod crustaceans: a review and prospectus. Journal of Experimental Marine Biology and Ecology 193:197–223.[CrossRef][Web of Science]
Alaska Fishery Research Bulletin Kimker A. (1994) Tanner crab survival in closed pots. 1:179–183.
Kondzela C. M. and Shirley T. C. (1993) Survival, feeding, and growth of juvenile Dungeness crabs from southeastern Alaska reared at different temperatures. Journal of Crustacean Biology 13:25–35.[CrossRef][Web of Science]
Miller R. J. (1978) Entry of Cancer productus to baited traps. Journal du Conseil International pour l'Exploration de la Mer 38:220–225.
Miller R. J. (1979) Saturation of crab traps: reduced entry and escapement. Journal du Conseil International pour l'Exploration de la Mer 38:338–345.
Miller R. J. (1990) Effectiveness of crab and lobster traps. Canadian. Journal of Fisheries and Aquatic Sciences 47:1228–1251.
Pearson W. H., Woodruff D. L., Sugarman P. C., Olla B. L. (1981) Effects of oiled sediment on predation on the littleneck clam, Protothaca staminea, by the Dungeness crab, Cancer magister. Estuarine, Coastal and Shelf Science 13:445–454.[CrossRef]
Rumble J. and Bishop G. (2002) Report to the Board of Fisheries, Southeast Alaska Dungeness Crab Fishery. Alaska Department of Fish and Game, Division of Commercial Fisheries Regional Information Report 1J02–45.
Fishery Bulletin US Shirley S. M. and Shirley T. C. (1988) Appendage injury in Dungeness crabs, Cancer magister, in southeastern Alaska. 86:156–160.
Simonson J. L. and Hochberg R. J. (1986) Effects of air exposure and claw breaks on survival of stone crabs Menippe mercenaria. Transactions of the American Fisheries Society 115:471–477.[CrossRef]
Smith B. D. and Jamieson G. S. (1989a) A model for standardizing Dungeness crabs (Cancer magister) catch rates among traps which experienced different soak times. Canadian Journal of Fisheries and Aquatic Sciences 46:1600–1608.
Smith B. D. and Jamieson G. S. (1989b) Exploitation and mortality of male Dungeness crabs (Cancer magister) near Tofino, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences 46:1609–1614.
Smith L. D. and Hines A. H. (1991) The effect of cheliped loss on blue crab Callinectes sapidus Rathbun foraging rate on soft-shell clams Mya arenaria L. Journal of Experimental Marine Biology and Ecology 151:245–256.[CrossRef][Web of Science]
Stewart J. E. (1993) Infectious diseases of marine crustaceans. In Couch J. A. and Fournie J. W. (Eds.). Pathobiology of Marine and Estuarine Organisms(CRC Press, Boca Raton, FL) pp. 319–342 565.
WDFW. (2005) Catch Record Cards. (Washington Department of Fish and Wildlife, Olympia, WA, 98501, USA) pp. 18 In 2005 Fishing in Washington Rule Pamphlet.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||