Seasonal, spatial, and postharvest variability in the survival of repeatedly discarded saucer scallops in Shark Bay, Western Australia.
KEY WORDS: fishery interaction, saucer scallop, Amusium balloti, mark-recapture, program MARK, CJS model
The saucer scallop Amusium balloti is a sedentary, short-lived (maximum life span, 3 y) species found in depths ranging between 15 m and 50 m in waters off the coasts of Western Australia and central Queensland (Dredge 2006), and in the lagoonal waters of New Caledonia and Chesterfield Reefs (Clavier 1991). In the Shark Bay region, this valuable but variable resource is captured using demersal otter trawl gear by 2 separately managed fishing fleets, the Shark Bay Scallop Managed Fishery that solely targets A. balloti, and the Shark Bay Prawn Managed Fishery that primarily targets king (Penaeus latisulcatus) and tiger (Penaeus esculentus) prawns, but also takes scallops as secondary target species (Kangas et al. 2011). Interfishery conflict (spanning the history of both fisheries) is an ongoing management challenge as the 2 industries clash over issues such as gear interactions, seasonal arrangements, equitable access to resource, and variable scallop recruitment. In response, a range of fishing strategies has been implemented primarily on a trial-and-error basis to mitigate and resolve these issues while maintaining sustainable fisheries. Periods of low scallop recruitment have triggered concerns about the role of regulatory discarding of scallops under past and present management measures by both fishing fleets. Understanding the drivers of the variable scallop recruitment patterns in Shark Bay is currently a research focus, including the impact from scallop discarding (Department of Fisheries 2010).
Survival of discards of target species from towed fishing gear is an important consideration in the overall management of a fishery resource (Davis & Olla 2001, Giomi et al. 2008), and one that has led to significant improvements in gear selectivity and postharvest procedures (Alverson et al. 1994, Kelleher 2005). Discard rates from scallop trawling or dredging can be high because scallop beds are targeted repeatedly until uneconomic harvest rates are reached (Maguire et al. 2002b). Discarded scallops can, therefore, be vulnerable to cumulative physical (Gilkinson et al. 1998, Schejter & Bremec 2007) and physiological injuries (Maguire et al. 2002a) where survival is highly dependent on the influences of various biological and environmental conditions concomitant with capture and processing techniques. Attempts to quantify discard mortality from short-term monitoring in the field or after a transfer to shore-based aquaria have produced numerous qualitative, behavioral, and physiological-based studies (Davis 2002, Broadhurst et al. 2006).
Saucer scallops in Shark Bay usually spawn during the winter/spring months (April through to December) followed by a 12-24-day larval period and settlement on the open sandy substrate of Shark Bay. Usually within 6-7 mo of spawning, juveniles reach 50-60 mm shell height by November and attain commercial sizes (>85 mm) by the following spawning season. Prior to 2004, when the prawn fishing season commenced (early March to April) before the scallop fishing season, it was mandatory for the prawn fleet to discard all retained scallops until the scallop fishery opened (often April or May). The later opening of the scallop fishery was to ensure that spawning stock abundance was adequate and that scallops were of marketable size and quality. From 2004 onward, both fleets began fishing simultaneously to prevent scallop discarding, but the prawn fleet is now required to discard all scallops during the key scallop spawning period (April to July) to optimize spawning success and recruitment settlement. Therefore, regulatory discarding of scallops occurred both during summer and winter months, but the discard rate is generally greater during the winter because of a higher abundance of small scallops (+0 age cohort), and also a greater proportion discarded by the prawn fleet because they use a smaller mesh size (50 mm) codend than the scallop fleet (100 mm). The impact of trawl disturbance and discarding on juvenile scallops may potentially be high.
Furthermore, postcapture methods and handling practices are additional factors that also affect the survival of discards (Bremec et al. 2004, Campbell et al. 2010). For example, one major change in the handling practices onboard prawn boats has been the introduction of seawater hoppers (large seawater tanks positioned on the back deck of trawlers). The benefits of a hopper system are in reducing the air exposure period, faster sorting and discarding, and an associated increase in the likelihood of survival in discards and bycatch (Heales et al. 2003), but this has not been demonstrated for Amusium balloti.
Survival estimates from repeated discarding and changes in postharvest methods are yet to be evaluated for trawl-captured saucer scallops in Shark Bay. This is a critical issue related to fishing fleet interactions because high discard mortality rates have the potential to affect scallop recruitment in addition to affecting adversely short-term catches in the fishery. In the current study, we investigated spatial and temporal differences in the short-term survival of discarded scallops under different postcapture treatments from field experiments simulating commercial scallop trawl activities. We hypothesized that the survival of discarded scallops would (1) be higher in winter than in summer, (2) be higher in prespawned scallops than in postspawned scallops, (3) be higher in hopper-treated scallops than in those sorted and handled in the open air (air exposed), and (4) demonstrate no differences among trawl sites. Because of the constraints of confining trawled animals to unnatural habitat conditions in aquaria experiments or release cages, and the logistical difficulties of working with such systems in very remote areas (Broadhurst et al. 2006), we adopted a tag--recapture approach to estimate survival and recapture rates of discarded scallops. Using this method, a proportion of the individually tagged and released scallops were recaptured and released repeatedly so scallops experienced the cumulative effects of discarding that is common under commercial practices. Individually identifiable scallops also provided quantitative information on recapture rates, recapture frequency, and recapture history during different fishing periods. Based on this information, the fishing impact of discarding (differently treated) scallops on the sustainability of the resource was evaluated under past and present management strategies for the Shark Bay trawl fisheries.
MATERIALS AND METHODS
Tag--recapture experiments were conducted within the Shark Bay region of western Australia between 25[degrees]00' S and 25[degrees]24' S, and 113[degrees]15' E and 113[degrees]25' E (Fig. 1). Sites of moderate scallop abundance (~3,000-5,000/trawl, 20-min shot) based on data from an annual scallop survey (undertaken in November and December of 2007 and 2008) were chosen for the tag-recapture experiments. Lower Denham Sound (LDS) and Upper Denham Sound (UDS) sites in the southern region of Shark Bay were selected for winter experiments whereas East Shark Bay (ESB) and West Shark Bay (WSB) sites in the central region of the bay were selected for summer experiments. Habitat differences among these sites are unknown, but generally the bottom consisted of sand and shell substrates with a trawl depth range of 15-25 m across all sites. Different regions within Shark Bay were chosen for winter and summer experiments on the basis of being no more than 60 min travel time between sites (to avoid operational fishing boats) and accessible during favorable weather conditions.
Scallop tagging and recapture experiments were conducted in winter (September 2008), when most scallops were in the postspawning phase, and in summer (February 2009), when scallops were in prespawning phase (Joll & Caputi 1995). On both occasions, the research vessel Naturaliste was configured to operate as a twin-gear-rigged otter trawl system (19.8 m headrope nets), using the standard 50-mm diamond mesh codend used by commercial prawn fishers. All experiments were conducted between the hours of 1830 and 0600 in accordance with commercial prawn fishing operations. Sea surface and air temperatures during winter experiments were 17-20[degrees]C and 19[degrees]C, respectively, with an average wind speed of 4.5 knots. During the summer, sea surface and air temperatures were 25-28[degrees]C and 25-30[degrees]C, respectively, with average wind speeds up to 4.8 knots.
Scallop Capture and Tagging Procedure
Experimental sites were trawled (20-min shot duration) to capture ~2,000 scallops for marking on the tagging night and kept in a recirculating seawater tank. For the air exposure treatment, batches of 200-250 scallops were removed from the holding tank at a time and were air exposed for approximately 40 min, during which time they were measured and tagged. A 40-min air exposure period was chosen because it is within the upper time limit scallops are handled before being shucked or discarded during commercial operations. For the hopper treatment, scallops were removed from the holding tank, one at a time, measured and tagged, and immersed in seawater again as fast as possible to minimize air exposure. Batches of tagged scallops from the hopper treatment were released together with tagged air-exposed scallops. Scallops were tagged with individually identifiable glue-on shellfish tags (Hallprint, FPN tags) attached with cyanoacrylate adhesive. It took approximately 4 h to complete the tagging and release of scallops. In addition, a control treatment tank (1 for each experiment) with flow-through seawater and containing approximately 300 trawl-captured, untagged scallops were kept onboard and undisturbed for 6-8 days to assess survival and condition of scallops that had experienced minimal trawl disturbance.
Experimental sites were trawled the following 4 nights in winter and 3 nights in summer (bad weather conditions precluded an additional trawl night in the summer experimental series) after the tagging night to recapture tagged scallops. A total of 7 trawl shots of each 10-min trawl duration were conducted over the experimental sites. Recaptured tagged scallops had their tag numbers recorded and were separated according to their respective treatment conditions. Air-treated scallops were kept in baskets on the sorting table whereas hopper-treated scallops were immersed in seawater. At the completion of all 7 shots, recaptured tagged scallops were released collectively on the surface as precisely as possible onto their capture location. Information on recovered dead, tagged scallop shells was retained as additional information but was not used in the survival analysis. Analyses of variance (ANOVAs) were used to compare differences in the size (shell height) of scallops between treatment groups and between captured and not recaptured scallops at each of the sampling sites using Statistica (version 7.1; Statsoft Inc., Tulsa, OK).
Damage and Adductor Muscle Weight Analyses
Direct damage and injury to scallops were assessed visually from a subsample of 300 scallops (tagged and untagged) collected across all trawl sites during the summer experiments using a damage scale from level 0 (no damage) to level 5 (dead scallop). Damage scale descriptions are as follows: level 0, no external damage or injury to valves or soft tissue; level 1, minor chipping to the edges of valves; level 2, major chipping to the edges of valves; level 3, extensive chipping of valves, exposing soft tissue; level 4, proportions of valves missing, visible injury to soft tissue but scallop alive; and level 5, valves cracked in half or smashed, resulting in death of scallop. Otter trawling for scallops relies on the animals swimming up into the water column to be captured, and therefore the capture of dead scallops in the nets was an unexpected result.
Maximum likelihood methods were used to estimate the conditional probabilities of apparent survival ([phi]) and resighting (p) of hopper-treated and air-exposed tagged scallops at the different sites during the winter and summer periods. Analyses of survival and resighting estimates were performed using the software program MARK developed by White and Burnham (1999). The modeling proccss started with assessing the goodness of fit of the data to the fully parameterized Cormack-Jolly-Seber (CJS) model with time (t = 1 to n repeat trawl nights) and group (g = hopper or air) dependence for survival and resighting ([[phi[.sub.g x t], [P.sub.g x t]). Information on recovered dead, tagged scallop shells was retained as additional information but was not used in the survival analyses. Goodness-of-fit testing of both winter and summer data sets detected overdispersion of the data and so the variance inflation factor ([??]) was calculated to adjust the sensitivity of the model selection process to the detection of fine-scale structural features within the data (Lebreton et al. 1992, Anderson et al. 1998). We chose to estimate [??] by dividing the observed model [??] by the mean d from a bootstrap analysis. Selection of the most parsimonious model from a candidate set of models under consideration was assessed by applying the quasilikelihood adjusted form of Akaike's information criterion ([QAIC.sub.c]) after incorporating d to allow for overdispersion of the data. If the [QAIC.sub.c] of the simplified model was lower than that of the starting model, the simplified model was adopted as the best general model. Model selection proceeded using ranked [QAIC.sub.c] values and expressed as [DELTA][QAIC.sub.c] calculated as the difference with the model with minimum [QAIC.sub.c]. These [DELTA][QAIC.sub.c] values were also used to compute model-specific [QAIC.sub.c] weights, which reflect the proportional support of evidence for a specific model. However, if [DELTA][QAIC.sub.c] was less than 2, then those models were considered to have equal likelihood of describing the data and thus model averaging was applied to derive estimates of survival and resighting probabilities (Burnham et al. 1995).
Although trawl sites were open to scallop movement, significant movement of released, tagged scallops was not considered to be an issue because Amusium balloti is a relatively sedentary species with limited capacity for movement (<20 m) (Joll 1989), and any active migration from the experimental sites was considered to be minimal. We also assumed tag loss and tag-induced mortality to be low and equivalent among all treatments at all sites. Assumptions relating to the standard CJS model were (1) every marked animal in the population has the same probability of recapture between trawl nights, (2) every marked animal has the same probability of surviving between trawl nights, (3) no tags are lost and tags are readable, and (4) all samples are instantaneous and each release occurs immediately after sampling
To determine the relative survival of hopper-treated scallops versus air-exposed scallops, data from scallops captured once only (and thus applied with the treatment once) was analyzed further using SURVIV, a software program that computes estimates of survival rates with multinomially distributed data (available free from http://www.mbr-pwrc.usgs.gov/software/surviv.shtml).
General Recapture Information
The total number of scallops tagged and recaptured for each treatment at each site during the experimental period is summarized in Table 1. Generally, recapture rates were higher (>40%) in winter than in summer treatments, with the highest number of scallops recaptured at UDS. There was also a trend of an increasing scallop recapture rate from nights 2-4 at UDS, whereas recapture rates at LDS were lower and more consistent. Conversely, in the summer recapture rates were less than 40% across all the nights at both sites. Recapture rates were higher at ESB for nights 1 and 2, with a rapid decline in recapture rates decreasing to less than 10% on night 3 at both ESB and WSB (Fig. 2). In winter, recapture rates from hopper-treated scallops were lower than air-exposed scallops at UDS, but the reverse trend was observed at LDS. In summer, hopper-treated scallops showed lower recapture rates than air-exposed scallops at both ESB and WSB (Fig. 2).
Recapture frequency of individually tagged scallops was highly variable across all sites and treatments. The capture rate of scallops caught once and twice only was lower at UDS than at LDS, but the capture rate of scallops caught 3 times and on all 4 nights was considerably higher at UDS than at LDS (Fig. 3). For summer, the capture rates of scallops caught once, twice, and 3 times were higher at ESB than at WSB. There were no clear, consistent patterns in the overall recapture frequency of individually tagged scallops between hopper and air-exposed treatments across all sites (Fig. 3). A small proportion of scallops recaptured in winter and summer were recovered dead, some with and some without soft tissue attached. The numbers of dead scallops recovered in summer were higher at ESB than at WSB and altogether considerably higher than in winter (Table 1). A total of 6 scallops were found dead at the end of the experimental period in the control treatment tanks in both winter and summer. This suggests that a 1-off trawl capture event may cause relatively low mortality.
Although a similar size range of scallops were tagged in winter (shell height, 78-111 mm) and summer (shell height, 50-112 mm), there were significant differences in the mean size of scallops tagged between sites in winter (UDS, 91.8 [+ or -] 0.3 mm; LDS, 89.1 [+ or -] 0.2 mm; [F.sub.1], 3,959 = 77.8, P < 0.01) and summer (ESB, 87.2 [+ or -] 0.2 mm; WSB, 86.0 [+or -] 0.1 mm; Fl, 3.953 = 28.2, P < 0.01). These significant differences are mostly the result of the large sample sizes, and the biological differences in terms of age or reproductive condition are not substantial. Thus, scallop size should not influence the tagging results significantly. There were no significant differences in the size of scallops between the hopper and air-exposed treatment groups and between recaptured and not recaptured scallops for the winter and summer periods (P > 0.05).
Goodness-of-Fit Testing Results
The results of the goodness-of-fit tests revealed overdispersion for the saturated models for all sites (P < 0.05). The poor fit ' indicated violations of CJS assumption of equal probability of capture and survival of tagged animals, which suggests hetero-geneity in survival or capture probabilities among individuals, or lack of independent sampling of individuals. The former is more likely based on our experimental design in which the reapplication of either the hopper or the air treatment on recapture meant not all tagged scallops experienced equal intensity of treatment effects during the experimental period. This was a deliberate component of our experimental design to simulate commercial settings in which some scallops experience cumulative effects of trawling, but it violates the assumption that every tagged animal has the same probability of recapture. Nonetheless, violation of the assumption does not negate a tag--recapture analysis because survival estimates from CJS models are robust to heterogeneity in data (Pollock et al. 1990). The results from SURVIV analyses showed the relative (ratio of hopper to air exposure) survival estimates to be close to 1.0 at all sites, indicating no significant differences in survival between treatments effects, thus corroborating the results from the MARK analyses. Because none of the candidate set of models weighted highly for treatment effects for either survival or recapture probabilities, we opted to proceed with the CJS model analyses and estimated d to be 3.82 for LDS, 4.84 for UDS, 2.42 for ESB, and 2.67 for WSB. We incorporated these values to calculate the [QAIC.sub.c] values and adjusted the SEs associated with parameter estimates.
Scallop Survival and Recapture Estimates
From the candidate set of models from the CJS analyses, resighting probability of discarded scallops was largely a function of time across all sites, but apparent survival varied by time, treatment group and was constant with period (Table 2). The best model for each site did not get strong support; the following 1-2 models were all plausible with similar likelihood and weighting ([DELTA][QAIC.sub.c] < 2), and so apparent survival and recapture rates were estimated using model averaging. Survival estimates in summer were highly variable between the 2 sites for all 3 nights. Apparent survival of scallops on the first night at WSB was 26% higher than at ESB. On the third night, however, survival was 20% higher at ESB than at WSB (Fig. 4A). Apparent survival on the second night was highest at ESB, with a range of 85-96%, and 78 % at WSB. However, large variation around these estimates suggests differences may not be significant between nights. Survival estimates in winter were greater than 90% for all 4 nights. Estimates for air-exposed scallops were higher than hopper-treated scallops at UDS but lower at LDS (Fig. 4B). The high variances around these estimates also suggest no marked differences in survival between sites or treatment groups in winter. Recapture probability estimates in summer decreased from 70% from the first night to 60% on the third night at WSB, whereas a greater decline occurred at ESB, where recapture probability estimates dropped from 85% to 55% in air-exposed scallops and to 48% in hopper-treated scallops (Fig. 4C). Recapture estimates in winter ranged between 45% and 55% across the 4 nights at LDS, whereas at UDS, estimates increased steadily from 55-84% on the fourth night (Fig. 4D).
The return rates (95% CI) of live scallops for each night of trawling are presented in Table 3. The average return rate across all nights in summer was 35%, whereas that for winter was 55%. The average return rates between air- and hopper-treated scallops were almost identical, with 48% for air and 47% for hopper treatments. There were spatial differences in return rates, with averages for LDS, UDS, ESB, and WSB being 47%, 64%, 34%, and 36%, respectively.
The majority of scallops sampled indicated level 1 injury, and 94% of assessed scallops were below a level 3 damage grading. Level 3 damage was seen in 4% of scallops, followed by 1.1% with level 4 damage and absolute mortality in 1.4% of scallops assessed (Fig. 5).
Survival Differences Among Discarded Amusium balloti
Large differences in the apparent survival of discarded scallops between the winter and summer periods highlight the potential significance of seasonal differences in environmental and/or physiological conditions in saucer scallops (Table 3). Desiccation is a major factor influencing the survival of scallops during emersion (Dickie 1958, Minchin et al. 2000, Jenkins & Brand 2001, Chen et al. 2007), and in the current study, scallops experienced mean air temperatures of 18[degrees]C in winter and 28[degrees]C in summer. One behavioral adaptation by scallops to minimize the effects of desiccation is to control valve gaping, to gain maximum benefit from gas diffusion and to optimize survival (Maguire et al. 2002a). However Amusium balloti cannot fully seal its shell opening and thus is highly susceptible to heat-related stress. Scallops attempting to conserve short-term energy would also switch to anaerobic respiration (Duncan 1993), but this can lead to death if the buildup of anaerobic metabolites is greater than the tolerance range of the animal (Christophersen et al. 2008). The rapid decline in recapture rate and recapture frequency over the 3 nights in the summer suggests that the cumulative metabolic stress from trawl activities was likely beyond its physiological threshold to cause death or to prolong the recovery phase. Cooler conditions in winter were likely to be more physiologically favorable, and reduced energetic demands during this period may have allowed scallops to sustain greater resilience to stress from the same trawl experience. Furthermore, saucer scallops can only be captured by trawling because they swim actively into the water column as an escape response (Joll 1989). Therefore, capture of dead scallop shells by trawl nets is an unusual occurrence. Despite this, high sighting of dead and moribund scallops during the summer suggests that a large number of discarded scallops had died directly from heat stress in summer, whereas others may succumb to ongoing mortalities caused by infection and predation (Jenkins et al. 2004, Himmelman et al. 2009). Summer air temperatures were likely within the upper thermal tolerance range of A. balloti.
The overall capture probabilities of discarded scallops were similar between hopper-treated and air-exposed discarded scallops (Table 3). The provision of a 40-min recovery period in water failed to elicit an improved capture rate in hopper-treated scallops versus those that were air exposed. If desiccation is a major stress factor in the survival of discarded scallops, it is unclear why postharvest treatment differences in survival were not apparent. It is possible that the 40-rain hopper treatment was an insufficient period for scallops to begin their recovery period, or the 24-h recovery period between retrawls was sufficient for the likelihood of survival from the 2 groups to become equivalent. In the absence of supportive data from behavioral or biochemical assays, it is difficult to assess quantitative differences in stress incurred by scallops from either treatment group. The 40-min treatment period applied in our study is the maximum exposure period during commercial operations, but Amusium balloti is reported to withstand exposure periods of up to 2 h (at 25[degrees]C) before experiencing appreciable mortality (Dredge 1997). The 24-h time interval between trawls and the 10-min trawl duration applied in our study are, however, not fully reflective of commercial practices, but rather were adopted for logistical reasons for the experiment. In Shark Bay, time intervals between trawls on the same grounds can vary from hours to days depending on other factors such as weather conditions, catch rates, bycatch, and fleet movement. The frequency of trawls on the same fishing ground is also highly variable, and discarded scallops may not be released on the same grounds where they are captured. Therefore, a full evaluation of hoppers on scallop survival would require testing a range of trawl durations and also a range of time intervals between trawls. Nonetheless, the benefits of hoppers include increased survival of invertebrate and soft-body bycatch organisms (Heales et al. 2003), aid in cleaning the grit from scallops, and retention of better meat condition during the sorting process (Chandrapavan, pers. obs.).
The physiological state of fatigued scallops and their ability to recover may also be a factor of their underlying reproductive condition, which is vastly different between the summer and winter periods. The scallop adductor muscle (harvested portion) has a dual role of mobilization of energy reserves in the form of glycogen during gametogenesis (winter months) and a primary role in fueling growth (summer months) and locomotion (Chantler 2006). In the current study, postspawned scallops in winter achieved greater survival and recapture estimates than prespawned scallops in summer, which suggests that reproductive investment by Amusium balloti did not influence stress recovery significantly. Campbell et al. (2010) also found inter- and intraseasonal differences in the survival of repeatedly trawled A. balloti, but these results were based on scallops being contained within catch bags and placed in trawl codends where scallops were not given the choice to respond independently to trawl gear. In undersize Pecten maxirnus, stress levels are known to vary by season with regard to the glycogen content driven by gonad development (Maguire et al. 2002c), whereas the tropical scallop Euvola ziczac shows decreased capacity to recover from exhaustive exercise following gonad maturation and spawning (Brokordt et al. 2000). However, reproductive energy demands during and after spawning did not affect significantly the escape response performance of adult Argopeeten purpuratus (Perez et al. 2009) or the scallop Placopecten magellanicus (Kraffe et al. 2008).
In the current study, tagging and recapturing scallops over a short time period allowed for the individual capture history of a discarded scallop to be documented. One interesting statistic from these data was that 92% of tagged scallops were recaptured at least once during the winter versus only 33% in summer, but of those recaptured, 85 % were resighted on the first night after release in the summer versus 61% in winter (Fig. 6). This suggests that although summer conditions are likely to be detrimental to discarded scallops, those that survive the initial release are able to recover faster because of their greater energy reserves, but are unlikely to recover from further cumulative trawl stress. This is supported by the presence of appreciable numbers of dead scallops during summer. In contrast, winter recapture rates increased with each night at 1 site and were stable at another. Reduced energy reserves of postspawned scallops in winter may inhibit a rapid swimming response to trawl disturbance whereas discarded scallops may not swim high enough into the water column to be captured; others may remain nonresponsive until they are fully recovered (Joll 1989). This would suggest seasonal differences in the catchability of discarded scallops where the physiological resilience to trawl-induced stress in Amusium balloti is dependent on its temperature tolerance range and available energy levels, and these aspects require further investigation.
Some scallops may have also succumbed to tag-induced mortality, but this is expected to be very low given the tagging process was noninvasive and trawl-captured scallops maintained within flow-through seawater tanks showed consistently low mortality rates (~3%) for both seasons. Direct damage to shell valves and soft tissue ranged from minor chipping, which can later appear as shock ring (growth scars) (Joll 1988), to irrevocable injury that results in immediate death. The current study found similar results to Amusium balloti harvested in the Queensland fishery Campbell et al. (2010), where estimates of dead scallops with crushed or cracked valves were very low (1%), and the majority incurred chipping to the outer edges of the valves. It was beyond the scope of this study to quantify other fates of discarded scallops (such as nonresponsive scallops to trawl nets and temporary emigration from trawl sites); however, the small number (<5 scallops) of tagged scallops that were not sighted during the experimental period but were captured several months later by commercial fishers does lend support to these possibilities.
Spatial differences in return rates of discarded scallops were evident both within and between the winter and summer periods, but this is unlikely to be significant given the high variances around the mean estimates. Overall capture probabilities were slightly higher at UDS than at LDS in winter, and higher at WSB than at ESB in summer (Table 3). Because different sites were chosen for the summer and winter periods, the interseasonal spatial variability in survival and recapture rates of discarded Amusium balloti is unclear. Given the close proximity of trawl sites within each season, it is also unlikely that oceanographic factors would have influenced return rates, although localized tidal movements may have played a role in shifting released scallops. Alternatively, it is possible that spatial differences in the quality of scallops (despite no significant differences in size) may have influenced their capture behavior. For example, the recapture rate of scallops caught 4 times in winter was approximately 20% greater at UDS than at LDS, whereas UDS was the only site where the percentage of scallops caught 4 times was greater than it being caught once (Fig. 3). Qualitative differences among scallops among different fishing grounds in Shark Bay does occur in terms of meat quality, where similar-size scallops show variation in meat size and weight (Sporer pers. comm.), and this may explain the small differences in the observed capture rates among sites.
Impact of Fishing Practices on Discarded Scallops
Marked seasonal differences in survival and recapture estimates of discarded Amusium balloti highlight the potential impact of past and present management strategies on the sustainability of the scallop resource in Shark Bay. Discarding during the warmer summer months, as was the case prior to 2004 by the prawn fleet, would have resulted in poor survival and a reduced catch and spawning biomass in later months. The change in regulation to simultaneous openings for both fleets (from 2004 onward) reflects a positive change in management strategy in which most scallops that are caught in summer are harvested. Scallops of nonmarket size (<85 ram), however, still continue to be discarded by both fleets and this requires either a change in product marketing for smaller size scallop meat or improved gear selectivity to reduce capture of sublegal-size scallops (Chandrapavan et al. 2012).
The forward shift in scallop season commencement in 2004 meant that the scallop fleet ceased fishing before peak spawning began, which resulted in an overall reduction in fishing intensity during the key scallop spawning months. The prawn fleet, however, continued its fishing operations during this period but were required to discard all scallops in their nets to maintain scallop abundance for spawning. Cooler conditions during the spawning months would favor greater survival of the discarded spawning scallops, whereas reproductive energy diverted to spawning is likely to prolong their recovery period, thus decreasing their catchability by trawl nets. To what extent the spawning behavior is altered as a result of trawl disturbance is unclear. For instance, it is not known whether trawl-induced stress has the potential to delay or hasten spawning, thus altering the natural timing of the spawning event. If larvae are produced (as a response to stress) when the environmental conditions are not optimal, then larval survival, movement, and settlement processes may be compromised, thus affecting the overall recruitment to the region. Similarly, trawling over newly settled scallops can be detrimental to their survival despite the seasonal advantage. Scallop spat and juveniles of size less than 30 mm in shell height are not readily retained by scallop or prawn nets, but are potentially vulnerable to gear damage as they pass through the fishing gear. Interactions with gear and with other scallops in the net are likely to result in shell damage because of their fragility, but the affect of trawl disturbance on the survival of juvenile scallops has not been investigated. This is another interfishery conflict issue between the Shark Bay trawl fisheries, and so the potential benefits of spatial closures in providing temporary protection to areas of high scallop settlement (before they reach subadult sizes) is under consideration as a supplementary management strategy.
Tag-recapture techniques have been successful in providing estimates of natural mortality (Naidu 1988), fishing mortality (Gruffydd 1972), and population sizes (Allison & Brand 1995) of major commercial scallop species. Indirect fishing mortality and delayed mortality estimates using tagging techniques are less common and mostly laboratory based. One of the major criticisms of both approaches is that estimates are determined under a limited range of fishing, environmental, and biological conditions and their interactions (Davis 2002). For tag--recaptures studies in particular, Broadhurst et al. (2006) commented on the high reliance of adequate tag returns (primarily from fishers) as a limitation of its methodology in providing reasonable survival estimates of discards. The current study attempted to address all these issues by incorporating differences in the seasonal, spatial, and post-harvest treatment of scallops into the field experiment to simulate a range of fishing conditions that scallops are likely to experience. Successive trawl surveys over a short time period ensured higher tag recapture rates, but, most important, allowed scallops to experience the cumulative stress effects from repeated trawling, thus providing more realistic estimates. Both Dredge (1997) and Campbell et al. (2010) tagged and released Amusium balloti after assessing their response to various postharvest treatments under simulated conditions (onboard tanks and in situ caging), and then relied on fisher tag returns to determine long-term survival estimates. In the current study, fishers were also encouraged (through a tag reward system) to target and return tagged scallops from the experimental sites, but fisher tag return rates were poor, with return rates of 2% and 21% from summer and winter sites respectively.
In conclusion, this study has provided estimates of survival and return rates of discarded scallops for a better understanding of the effect of discarding under different fishing conditions. For the management of the scallop resource, the study highlights the importance of minimizing discarding as much as possible during the summer months and the protection from trawl disturbance for scallops during the winter months. Achieving these objectives could potentially minimize some of the interfishery conflict and could promote scallop recruitment in Shark Bay.
We thank D. Boddington, S. Brown, M. Shanks, and N. Shaw for field assistance, and thank the skipper and crew of the Naturaliste. Special thanks to K. Pollock for statistical advice and assistance with MARK program analyses. This research was funded jointly by FRDC (project no. 2007/051) and the Shark Bay prawn and scallop licensees.
Allison, E. H. & A. R. Brand. 1995. A mark-recapture experiment on queen scallops, Aequipecten opercularis, on a North Irish Sea fishing ground. J. Mar. Biol. Assoc. UK 75:323-335.
Alverson, D. L., M. H. Freeberg, S. A. Murawski & J. G. Pope. 1994. A global assessment of fisheries bycatch and discards. FAO (United Nations Food and Agriculture Organization) fisheries technical paper no. 339. FAO, United Nations, Rome, Italy, 233 p.
Anderson, D. R., K. P. Burnham & G. White. 1998. Comparisons of Akaike information criterion for model selection and statistical inference from capture-recapture studies. J. Appl. Stat. 25:263 282.
Bremec, C. S., M. L. Lasta & D. Hernandez. 2004. Survival of Patagonian scallop (Zygochlamys patagonica, King and Broderip, 1832) after the size selection process on commercial fishing vessels. Fish. Res. 66:49-52.
Broadhurst, M. K., P. Suuronen & A. Hulme. 2006. Estimating collateral mortality from towed fishing gear. Fish Fish. 7:180-218.
Brokordt, K. B., J. H. Himmelman, O. A. Nusetti & H. E. Guderley. 2000. Reproductive investment reduces recuperation from exhaustive escape activity in the tropical scallop Euvola zizac. Mar. Biol. 137:857-865.
Burnham, K. P., G. C. White & D. R. Anderson. 1995. Model selection in the analysis of capture-recapture data. Biometrics 51:888-898.
Campbell, M. J., A. B. Campbell, R. A. Officer, M. F. O'Neill, D. G. Mayer, A. Thwaites, E. J. Jebreen, A. J. Courtney, N. Gribble, M. L. Lawrence, A. J. Prosser & S. L. Drabsch. 2010. Harvest strategy evaluation to optimise the sustainability and value of the Queensland scallop fishery. FRDC report 2006/024. Department of Employment, Economic Development and Innovation (DEEDI), Queensland, Australia, 141 pp.
Chandrapavan, A., M. Kangas & E. Sporer. 2012. Performance of sqaure-mesh codends in reducing discards and by-catch in the Shark Bay scallop fishery. Mar. Fresh. Res. 63:1-10.
Chantler, P. D. 2006. Scallop adductor muscle: structure and function. In: E. E. Shumway & G. J. Parsons, editors. Scallops: biology, ecology and aquaculture. Developments in aquaculture and fisheries science, vol. 35. Sydney: Elsevier. pp. 229-316.
Chen, M. Y., H. S. Yang, M. Delaporte & S. J. Zhao. 2007. Immune condition of Chlamys farreri in response to acute temperature challenge. Aquaculture 271:479-487.
Christophersen, G., G. Roman, J. Gallagher & T. Magnesen. 2008. Post-transport recovery of cultured scallop (Pecten maximus) spat, juveniles and adults. Aquacult. Int. 16:171-185.
Clavier, J. 1991. Etat des connaissances sur Amusium balloti (Bivalve, Pectinide) darts les lagons de Nouvelle Caledonie. Convention Sciences de la mer. Biologie Marine No 4. 54 pp.
Davis, M. W. 2002. Key principles for understanding fish by catch discard mortality. Can. J. Fish. Aquat. Sci. 59:1834-1843.
Davis, M. W. & B. L. Olla. 2001. Stress and delayed mortality induced in Pacific halibut by exposure to hooking, net towing, elevated seawater temperature and air: implications for management of bycatch. North Am. J. Fish. Manage. 21:725 732.
Department of Fisheries. 2010. Shark Bay prawn and scallop fisheries: final review report. Fisheries management paper no. 235. Western Australia. Perth, Department of Fisheries, Western Australia, 148 pp.
Dickie, L. M. 1958. Effects of high temperature on survival of the giant scallop. J. Fish. Res. Board Can. 15:1189 1211.
Dredge, M. C. L. 1997. Survival of saucer scallops, Amusium japonicum balloti, as a function of exposure time. J. Shellfish Res. 16:63-66.
Dredge, M. C. L. 2006. Scallop fisheries, mariculture, and enhancement in Australia. In: E. E. Shumway & G. J. Parsons, editors. Scallops: biology, ecology and aquaculture. Developments in aquaculture and fisheries science, vol. 35. Sydney: Elsevier. pp. 1391-1412.
Duncan, P. F. 1993. Post-harvest physiology of the scallop Pecten-maximus (L.). PhD diss., University of Glasgow. 184 pp.
Gilkinson, K., M. Paulin, S. Hurley & P. Schwinghamer. 1998. Impacts of trawl door scouring on infaunal bivalves: results of a physical trawl door model dense sand interaction. J. Exp. Mar. Biol. Ecol. 224:291-312.
Giomi, F., S. Raicevich, O. Giovanardi, F. Pranovi, P. Di Muro & M. Beltramini. 2008. Catch me in winter! Seasonal variation in air temperature severely enhances physiological stress and mortality of species subjected to sorting operations and discarded during annual fishing activities. Hydrobiologia 606:195-202.
Gruffydd, L. L. 1972. Mortality of scallops on a Manx scallop bed due to fishing. J. Mar. Biol. Assoc. UK 52:449-455.
Heales, D. S., D. T. Brewer & P. N. Jones. 2003. Subsampling trawl catches from vessels using seawater hoppers: are catch composition estimates biased? Fish. Res. 63:113-120.
Himmelman, J. H., H. E. Guderley & P. F. Duncan. 2009. Responses of the saucer scallop Amusium balloti to potential predators. J. Exp. Mar. Biol. Ecol. 378:58-61.
Jenkins, S. R. & A. R. Brand. 2001. The effect of dredge capture on the escape response of the great scallop, Pecten maximus (L.): implications for the survival of undersized discards. J. Exp. Mar. Biol. Ecol. 266: 33-50.
Jenkins, S. R., C. Mullen & A. R. Brand. 2004. Predator and scavenger aggregation to discarded by-catch from dredge fisheries: importance of damage level. J. Sea Res. 51:69-76.
Joll, L. M. 1988. Daily growth rings in juvenile saucer scallops, Amusium balloti (Bernardi). J. Shellfish Res. 7:73-76.
Joll, L. M. 1989. Swimming behavior of the saucer scallop Amusium balloti (Mollusca, Pectinidae). Mar. Biol. 102:299-305.
Joll, L. M. & N. Caputi. 1995. Geographic variation in the reproductive cycle of the saucer scallop, Amusium balloti (Bernardi, 1861) (Mollusca: Pectinidae), along the western Australian coast. Mar. Freshw. Res. 46:779-792.
Kangas, M., E. Sporer, S. Brown, M. Shanks & A. Chandrapavan & A. Thomson. 2011. Stock assessment for the Shark Bay Scallop Fishery. Fisheries research report no. 226. Department of Fisheries. Perth, Western Australia. 76 pp.
Kelleher, K. 2005. Discards in the world's marine fisheries: an update. FAO fisheries technical paper 470. Rome: Food and Agriculture Organization of the United Nations. 131 pp.
Kraffe, E., R. Tremblay, S. Belvin, J. R. LeCoz, Y. Marty & H. Guderley. 2008. Effect of reproduction on escape responses, metabolic rates and muscle mitochondrial properties in the scallop Placopecten magellanicus. Mar. Biol. 156:25-38.
Lebreton, J., K. P. Burnham, J. Clobert & D. R. Anderson. 1992. Modelling survival and testing biological hypotheses using marked animals: a unified approach with case studies. Ecol. Monogr. 62:67-118.
Maguire, J. A., A. Coleman, S. Jenkins & G. M. Burnell. 2002a. Effects of dredging on undersized scallops. Fish. Res. 56:155-165.
Maguire, J. A., S. Jenkins & G. M. Burnell. 2002b. The effects of repeated dredging and speed of tow on undersized scallops. Fish. Res. 58:367-377.
Maguire, J. A., M. O'Donoghue, S. Jenkins, A. Brand & G. M. Burnell. 2002c. Temporal and spatial variability in dredging induced stress in the great scallop Pecten maximus (L.). J. Shellfish Res. 21: 81-86.
Minchin, D., G. Haugum, H. Skjaeggestad & O. Strand. 2000. Effect of air exposure on scallop behaviour, and the implications for subsequent survival in culture. Aquacult. Int. 8:169-182.
Naidu, K. S. 1988. Estimating mortality rates in the Iceland scallop, Chlamys islandica (O.F. Muller). J. Shellfish Res. 7:61-71.
Perez, H. M., K. B. Brokordt, G. Martinez & H. Guderley. 2009. Locomotion versus spawning: escape responses during and after spawning in the scallop Argopeeten purpuratus. Mar. Biol. 156:1585-1593.
Pollock, K. H., J. D. Nichols, C. Brownie & J. E. Hines. 1990. Statistical inference for capture-recapture experiments. Wildl. Monogr. 107:1-97.
Schejter, L. & C. Bremec. 2007. Repaired shell damage in the commercial scallop Zygochlamys patagonica (King & Broderip, 1832). Argentine Sea. J. Sea. Res. 58:156-162.
White, G. C. & K. P. Burnham. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study 46:120-139.
ARANI CHANDRAPAVAN,* MERVI I. KANGAS AND ERROL C. SPORER
Western Australian Fisheries and Marine Research Laboratories, PO Box 20, North Beach, Western Australia, 6920, Australia
* Corresponding author. E-mail: Arani.Chandrapavan@fish.wa.gov.au
TABLE 1. Summary of total number of scallops tagged, recaptured, not recaptured, and recovered dead during the summer and winter experiments. Season [right arrow] Winter Sites [right arrow] Lower Denham Sound Treatment [right arrow] Hopper Air exposed Total tagged 940 944 Recaptured on night 1 504 459 Recaptured on night 2 400 359 Recaptured on night 3 405 357 Recaptured on night 4 420 404 Not recaptured 116 168 Total dead recoveries 2 5 Mean size [+ or -] SE (SH) Recaptured scallops 89.5 [+ or -] 0.2 89.3 [+ or -] 0.2 Not recaptured scallops 85.6 [+ or -] 1.4 87.8 [+ or -] 0.5 Season [right arrow] Winter Sites [right arrow] Upper Denham Sound Treatment [right arrow] Hopper Air exposed Total tagged 787 787 Recaptured on night 1 467 484 Recaptured on night 2 420 411 Recaptured on night 3 476 510 Recaptured on night 4 559 602 Not recaptured 70 44 Total dead recoveries 4 3 Mean size [+ or -] SE (SH) Recaptured scallops 91.5 [+ or -] 0.2 91.4 [+ or -] 0.2 Not recaptured scallops 86.2 [+ or -] 2.0 87.5 [+ or -] 2.3 Season [right arrow] Summer Sites [right arrow] East Shark Bay Treatment [right arrow] Hopper Air exposed Total tagged 951 950 Recaptured on night 1 195 209 Recaptured on night 2 109 147 Recaptured on night 3 50 65 Recaptured on night 4 N/A N/A Not recaptured 737 738 Total dead recoveries 54 42 Mean size [+ or -] SE (SH) Recaptured scallops 87.6 [+ or -] 0.4 86.6 [+ or -] 0.5 Not recaptured scallops 87.6 [+ or -] 0.3 87.0 [+ or -] 0.3 Season [right arrow] Summer Sites [right arrow] West Shark Bay Treatment [right arrow] Hopper Air exposed Total tagged 1,000 991 Recaptured on night 1 353 340 Recaptured on night 2 294 327 Recaptured on night 3 62 89 Recaptured on night 4 N/A N/A Not recaptured 517 513 Total dead recoveries 23 36 Mean size [+ or -] SE (SH) Recaptured scallops 86.0 [+ or -] 0.3 86.4 [+ or -] 0.3 Not recaptured scallops 85.4 [+ or -] 0.3 86.2 [+ or -] 0.3 Mean size (SH in millimeters) of recaptured and not recaptured scallops across all treatments and sites are also indicated. SH, shell height. TABLE 2. Cormack-Jolly-Seber models best fitting the data ([DELTA][QAIC.sub.c], < 2) to provide estimates of survival (o) and recapture (p) probabilities as a function of treatment group (g) or time (t) across all experimental sites. [DELTA] [QAIC.sub.c] Model [QAIC.sub.c] [QAIC.sub.c] weight Winter e(g) p(t) 2,668.96 0.00 0.32 LDS e(.) p(t) 2,669.27 0.31 0.27 Winter o(.) p(t) 1,692.01 0.00 0.38 UDS e(t) p(t) 1,692.54 0.53 0.29 o(g) p(t) 1,692.58 0.57 0.29 Summer o(t) p(g x t) 1,386.59 0.00 0.40 ESB o(g x t) p(t) 1,387.02 0.43 0.32 o(t) p(t) 1,387.99 1.40 0.20 Summer e(t) p(.) 2,016.14 0.00 0.38 WSB e(t) p(t) 2,016.72 0.58 0.28 o(t) p(g) 2,017.96 1.83 0.15 Model Parameters, Model likelihood n Winter e(g) p(t) 1.00 6 LDS e(.) p(t) 0.86 5 Winter o(.) p(t) 1.00 5 UDS e(t) p(t) 0.77 7 o(g) p(t) 0.75 6 Summer o(t) p(g x t) 1.00 8 ESB o(g x t) p(t) 0.81 8 o(t) p(t) 0.50 5 Summer e(t) p(.) 1.00 4 WSB e(t) p(t) 0.75 5 o(t) p(g) 0.40 5 ESB, East Shark Bay; LDS, Lower Denham Sound; QAIC,, Akaike's information criteria; UDS, Upper Denham Sound; WSB, West Shark Bay. TABLE 3. Estimates of return rates (95% CI) of live scallops computed as the product of survival and recapture probability model estimates (op) for each night of repeat trawling by the different postcapture treatment effect at the different sites during the winter and summer periods. Retrawl Treatment night Winter, LDS Winter, UDS Hopper Night 1 0.50 (0.42-0.56) 0.58 (0.50-0.65) Night 2 0.43 (0.34-0.52) 0.54 (0.45-0.61) Night 3 0.46 (0.37-0.53) 0.65 (0.54-0.72) Night 4 0.50 N/A 0.79 N/A Air Night 1 0.48 (0.40-0.56) 0.59 (0.50-0.66) Night 2 0.42 (0.34-0.50) 0.54 (0.45-0.61) Night 3 0.45 (0.35-0.52) 0.65 (0.55-0.72) Night 4 0.49 N/A 0.80 N/A Retrawl Treatment night Summer, ESB Summer, WSB Hopper Night 1 0.20 (0.15-0.26) 0.35 (0.28-0.41) Night 2 0.38 (0.09-0.62) 0.57 (0.39-0.72) Night 3 0.25 N/A 0.19 N/A Night 4 Air Night 1 0.20 (0.15-0.26) 0.35 (0.28-0.41) Night 2 0.51 (0.05-0.67) 0.58 (0.39-0.73) Night 3 0.29 N/A 0.20 N/A Night 4 ESB, East Shark Bay; LDS, Lower Denham Sound; N/A, not applicable/ UDS, Upper Denham Sound; WSB, West Shark Bay.
|Printer friendly Cite/link Email Feedback|
|Author:||Chandrapavan, Arani; Kangas, Mervi I.; Sporer, Errol C.|
|Publication:||Journal of Shellfish Research|
|Date:||Dec 1, 2012|
|Previous Article:||Physiological effects and biotransformation of PSP toxins in the New Zealand scallop, Pecten novaezelandiae.|
|Next Article:||Use of length-frequency analysis for growth estimation of the California two-spotted octopus Octopus bimaculatus Verril 1883 of the Gulf of...|