Marine Pollution Bulletin 79 (2014) 261–267

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Ghost fishing activity in derelict blue crab traps in Louisiana Julie A. Anderson a,b,⇑, Amy B. Alford a a b

Louisiana Sea Grant College Program, Louisiana State University, Baton Rouge, LA 70803, USA School of Renewable Natural Resources, Louisiana State University Agricultural Center, Baton Rouge, LA 70803, USA

a r t i c l e

i n f o

Keywords: Callinectes sapidus Louisiana blue crab fishery Derelict crab traps Citizen scientists Ghost fishing Marine debris

a b s t r a c t Derelict crab traps impact the coastal ecosystem through continued catch of target species and species of conservation, economic, or recreational importance. During volunteer-supported crab trap cleanups in 2012 and 2013, we quantified ghost fishing activity in derelict crab traps in coastal Louisiana through a citizen scientist program. Volunteers removed 3607 derelict traps during these events, and over 65% of traps analyzed by citizen scientists were actively ghost fishing. Additionally, volunteers identified 19 species enmeshed in derelict traps, including a combination of fresh and saltwater species. We also detected a significant difference in the number of blue crab in actively ghost fishing derelict traps across removal locations with estimated catches varying between 2.4 and 3.5 crabs/trap. Our instantaneous estimates of ghost fishing activity are greater than those previously thought in Louisiana, further justifying current derelict crab trap prevention and removal extension and outreach programs in Louisiana and throughout the Gulf of Mexico. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction The introduction of wire traps in the mid-20th century greatly increased the commercial and recreational harvest of blue crab (Callinectes sapidus, Rathbun, 1896) in the Gulf of Mexico (Guillory, 1993; Guillory et al., 2001b). However the increased use of new gear has resulted in derelict (i.e., no longer tended to by fishermen) gear such as traps, lines, and buoys contributing to the problem of marine debris (Coe and Rogers, 1997; Good et al., 2010; Havens et al., 2008). Blue crab traps become derelict through a variety of means: storm activity, use of inferior rope and floats, vandalism, improper disposal, and vessel propellers severing buoy lines (Havens et al., 2008). The derelict gear can pose a variety of harms to the coastal ecosystem through continued catch of the target species (Breen, 1987; Havens et al., 2011), bycatch of species of conservation, economic, or recreational importance (Udyawer et al., 2013), and user group conflicts (Havens et al., 2011). For many crustacean species including blue crab, capture in derelict trap can result in increased predation, cannibalism, starvation, loss of appendage, and a general decline in health (see review in Maselko et al. (2013)). Whereas derelict crab traps are no longer baited by fishermen, continual mortality of organisms within these traps causes an increase in fishing mortality through the phenomenon of ghost fishing with dead organisms then serving to attract ⇑ Corresponding author at: School of Renewable Natural Resources, Louisiana State University Agricultural Center, Baton Rouge, LA 70803, USA. Tel.: +1 225 578 0771; fax: +1 225 578 4227. E-mail address: [email protected] (J.A. Anderson). 0025-326X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.marpolbul.2013.12.002

more individuals (Breen, 1987). Additionally, if empty, organism may seek shelter in a trap and may become entrapped resulting in ghost fishing. In Louisiana, estimates of ghost fishing provided by Guillory (1993) suggested that each derelict trap contributed to the mortality of 26 blue crabs annually. When these mortality estimates are extrapolated to the number of traps lost annually in Louisiana coastal waters (e.g., 257 per fisherman; Guillory et al., 2001a), a significant effect on the blue crab fishery is realized with approximately 1816 active commercial crab fishermen, (Louisiana Department of Wildlife and Fisheries, 2011; West et al., 2011). Consequently, 12 million crabs or 2 million kilograms of potential harvest may be lost to ghost fishing mortality representing 10% of the annual landings reported in Louisiana in 2012 or about US$4 million (Louisiana Department of Wildlife and Fisheries, 2013). Additionally, in 2012, approximately 6000 recreational crab trap licenses were sold to Louisiana residents, and no estimates exist on loss rate or potential ghost fishing impact related to recreational trapping (Louisiana Department of Wildlife and Fisheries, 2012). While the effect has not been quantified in Louisiana, other marine species of recreational, commercial, and conservation importance also endure problems associated with derelict traps. Fish and invertebrate species such as spotted sea trout (Cynoscion nebulosus), red drum (Sciaenops ocellatus), and stone crab (Menippe adina) have been observed in derelict crab traps as well as diamond back terrapins (Malaclemys terrapin) and river otter (Lontra canadensis) (Guillory et al., 2001b; Harden and Williard, 2012; Havens et al., 2008; Radzio et al., 2013).

262

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267

Traps caught in shrimp fishing gear can clog turtle excluder devices preventing catch of shrimp and escape of turtles, and this furthers conflicts between shrimp and crab fishermen (recreational and commercial) (Guillory et al., 2001a,b). Additionally, derelict traps serve as an underwater hazard for boaters especially if the buoy line has been severed leading to additional economic losses (McIlgorm et al., 2011). Although design modifications (e.g., biodegradable panels; Bilkovic et al., 2012) and best management practices related to gear maintenance have been suggested to reduce the impact of derelict traps on marine ecosystems, removal of derelict traps is the only absolute solution. Since 2004, the Louisiana Department of Wildlife and Fisheries (LDWF) has directed largely volunteersupported derelict trap removal efforts through the authority granted by Louisiana R.S. 56:322 (N). In Louisiana, derelict trap removal by anyone other than the owner can only occur during a specified 10 day fishery closures between February 1 and March 31. During these closures, designated areas are legally closed to crab fishing, and all traps remaining in these zones are considered derelict and subject to removal. However, volunteer support and therefore effective derelict trap removal efforts declined by 2009, and as such LDWF partnered with the Louisiana Sea Grant College Program (LSG) in 2012 to create a new cleanup event to encourage greater volunteer participation called the ‘Crab Trap Rodeo.’ In addition to augmenting the removal of derelict crab traps, the increased volunteer support at these Rodeos created a unique opportunity to have citizen scientists collect ghost fishing data from derelict crab traps in 2012 and 2013. Citizen scientists have participated in other marine debris projects quantifying types, impacts, and other aspects of marine debris (Hidalgo-Ruz and Thiel, 2013; Ribic et al., 2012). Utilizing volunteers to act as citizen scientists is beneficial in many respects. Primarily, using volunteers as citizen scientists allows for much more information to be collected when resources are limited (Bonney et al., 2009; Hidalgo-Ruz and Thiel, 2013; Silvertown, 2009). Because derelict traps can only be removed during the closures, time is a limiting factor with respect to data collection. However, because the overall purpose of the closures, and therefore, the volunteer supported cleanup events is to remove as many traps from Louisiana’s waterways as possible, the opportunity exists to collect a large amount of data related to derelict crab trap prevalence and ghost fishing activity. Additionally, the incorporation of citizen scientists in the derelict crab trap removal events engages concerned volunteers and encourages environmental stewardship. Therefore, our goal was to quantify ghost fishing in derelict crab traps during trap removal events (i.e., Rodeos) using citizen scientists and scientist volunteers. The information collected will help guide blue crab fishery management decisions in Louisiana. Our objectives were to (1) describe the condition of derelict traps in Louisiana, (2) provide baseline data regarding the species captured by derelict traps, (3) provide estimates of the ghost fishing rates of derelict traps, and (4) provide recommendations for locations of future derelict crab trap removal events given specific management or conservation goals.

2. Methods 2.1. Study location and design The locations of the 2012 and 2013 blue crab fishery closures were predetermined by LDWF personnel during a pre-assessment of areas based on need and feasibility (e.g., access and shallow water allowing removal of visible traps). Closures had clear GPS boundaries, and maps were provided to all volunteers. Blue crab fishery closure areas encompassed 2291 km2 of typical cord grass

(Spartina spp.) dominated salt marsh habitats in southeastern Louisiana (Table 1). Specifically, closures in 2012 covered the marshes north of Breton Sound (Delacroix closure; Fig. 1) and Terrebonne Bay (Cocodrie closure) while closures in 2013 covered Breton Sound (Pointe a la Hache closure) and the marshes south of Lake Borgne (Hopedale closure). Rodeo dates were selected by LSG based on feasibility within the February 1–March 31 allowable timeframe and low tides making many trap visible. Rodeos were held on 25 February, 3 March, and 17 March 2012 and 16 February, 23 February, and 9 March 2013 from 0830 h. to 1600 h. Volunteers were recruited through regional outreach events and radio and print advertisements. Although Rodeos were held on Saturdays to encourage greater volunteer participation rates, removal of derelict traps continued throughout the closure periods and was primarily facilitated by LDWF personnel and anonymous citizens. Volunteers were instructed to remove all visible derelict traps and to occasionally dredge the seafloor when possible. All traps returned to the collection sites were crushed with a hydraulic press and sent to appropriate disposal and recycling facilities per Louisiana R.S. 56:322 (N). 2.2. Sampling protocol According to Bonney et al. (2009), obtaining practical citizen scientist data requires providing a clear protocol, simple and easy to understand data collection forms, and support. The protocol we provided to citizen scientists was designed with consultation from other state agencies that have conducted derelict crab trap removal programs and various participants. We reduced the extensive list of potential variables related to derelict trap condition and ghost fishing activity such that citizen scientists only had to complete one simple data collection form per trap (Cohn, 2008). One form per trap was determined to be the best for volunteers based on pilot trials with undergraduate students and prioritized the collection of the most relevant data. Quality control is also an important aspect of citizen scientist programs (Hidalgo-Ruz and Thiel, 2013). Because volunteers were aboard watercraft far from shore during data collection and all live animals recorded as enmeshed in traps were released when derelict traps were removed from the marsh, we initially asked volunteers to also take photos of each trap in anticipation that we would use these photos as a quality control check. Additionally, in both years volunteers brought back some species to the dock for assistance in species identification. As feasible, we recruited one volunteer on each boat to record data from derelict traps that were removed by the team of volunteers on the boat. Every cleanup team did not collect data as the overall goal of the event was to remove traps. For willing volunteers, we provided a binder containing data collection forms and a species identification chart. We instructed volunteers how to record trap integrity (e.g., intact or collapsed), mesh type (square or hexagonal), and number of individuals (live or dead) of each species encountered in each derelict trap. If clear identification of dead individuals was not possible, volunteers were instructed to identify the carcasses to the lowest taxonomic level possible (e.g., bird, mammal). We instructed volunteers to release all live organisms immediately after data collection from each derelict trap. We collected datasheets as boats returned to the trap collection site. Onshore, we recruited a volunteer to count all traps that came off boats. 2.3. Data analysis We summarized data collected from derelict traps for each Rodeo. If the trap had at least one live or dead enmeshed organism, we designated a trap as ghost fishing. We evaluated the differences in the proportion of ghost fishing traps across locations and mesh

263

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267 Table 1 Summary of Derelict Crab Trap Rodeos at blue crab fishery closure locations in southeastern Louisiana during late winter 2012–2013. Year

Location a

Date

km2

Volunteers

Boats

Traps removed (Ni)

145 64 95

27 13 23

1435 522 747

62 37 89

10 5 15

345 147 411

2012

Delacroix Delacroix Cocodrie

25 February 3 March 17 March

947

2013

Pointe a la Hachea Pointe a la Hache Hopedale

16 February 23 February 9 March

503

237

604

a

Data from the first Rodeos in 2012 and 2013 include traps collected by LDWF personnel, volunteers, and anonymous fishermen during the week between the first and second Rodeos held at the same location.

Fig. 1. Location of blue crab fishery closure areas and Rodeo sites. Locations are (A) Cocodrie; (B) Pointe a la Hache; (C) Delacroix; and (D) Hopedale.

types in separate Chi-squared tests (a = 0.05) in the stats package in Program R ver. 3.0.0 (R Development Core Team, 2013). To compare the number of blue crab enmeshed in derelict traps across Rodeo locations, we used a zero-inflated Poisson model (ZIP; a = 0.05) with a log link function for non-zero observations and a logit link function for binary data in the pscl package in Program R. We inspected the frequency distribution of the count data and found the distribution to be heavily influenced by a high frequency of zeros. Even if a derelict trap is present, it does not necessary actively ghost fish. For example, derelict traps may be discarded above the water line or may have non-functional entrance funnels and therefore, not able to trap blue crabs. Additionally, we conducted goodness-of-fit tests and found the data did not fit a normal or Poisson distribution and therefore determined the ZIP model to be more appropriate to realistically estimate blue crab ghost fishing by derelict traps (Martin et al., 2005). The use of approaches that assume normal distributions (e.g., t-tests) would inaccurately describe and compare the number of blue crab enmeshed in each actively ghost fishing derelict trap. The zero-inflated Poisson model allowed us to simultaneously estimate the number of blue crab in each ghost fishing trap at each closure area ð^ ki Þ given a Poisson distribution and the probability that a trap would have no enmeshed ^ð0Þi Þ under a binomial distribution (Martin et al., 2005). crab ðp Finally, we used modified equations from Maselko et al. (2013)

to estimate the total number of blue crab released from ghost fishing traps at each location, Tbi as

^ð0Þi ÞÞ Tb i ¼ N i  ^ki  ð1  p where Ni the total number of traps is removed during the closure ^ð0Þi were estimated from period at area i. The parameters ^ ki and p the ZIP model. We estimated 95% confidence bounds for Tbi from a bootstrap distribution of 1000 samples with replacement from the sample of derelict traps analyzed for ghost fishing activity by the citizen scientists at each area (ni) in the boot package in Program ^ð0Þi were recalculated for each subsample R. Parameters ^ ki and p and were approximated as

Xg k0i

¼

t¼1

crabti

gi

And

p0 ð0Þi ¼ 1 

gi ni

where crabti is the number of crab caught in a ghost fishing trap t at area i and gi is the number of traps that had at least 1 enmeshed crab at area i.

264

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267

3. Results

100%

Volunteers and personnel removed 2704 derelict traps during the closures in 2012 and 903 traps during the closures in 2013 (Table 1). Volunteers provided ghost fishing data for 1073 traps; 27 citizen scientist teams provided completed data collection forms for 380 traps from five Rodeo events. Because data collected from traps during the two Rodeo events at Delacroix in 2012 covered the same 10 day closure, we combined the data from these two days as one event. Although derelict trap removal continued, data were not collected from derelict crab traps during the 23 February 2013 Rodeo event at Pointe a la Hache due to inclement weather. Volunteers identified 19 species captured in 1073 traps (Table 2). Oysters were found on traps at all locations. Although all species found enmeshed in derelict traps are known to occasionally inhabit estuarine habitats, we further classified invertebrate and fish species as ‘‘freshwater’’ or ‘‘marine’’ according to primary habitat occurrences (Christmas, 1973; Ross, 2001). Delacroix had the greatest number of species and the greatest number of freshwater fish species (Table 2). Of the species other than blue crab found in derelict traps, marine fish species, stone crabs, and diamondback terrapins were the most abundant bycatch (Fig. 2). Of the traps with complete data collection forms (n = 380), a majority of the traps (71%; Table 3) were constructed of square mesh and many traps (72%) were still intact and capable of fishing (i.e., all six side panels of a trap were connected). Sixty-seven percent of the total traps removed were actively ghost fishing any species. The proportion of intact derelict traps that were actively ghost fishing did not differ significantly between mesh types (v2 = 0.343, df = 1, P = 0.558). However, the proportion of derelict traps that were ghost fishing did differ significantly across Rodeo locations (v2 = 39.2817, df = 3, P < 0.0001) with Hopedale having the greatest proportion of traps that were ghost fishing any species (Table 3). Citizen scientists released 671 live blue crabs and 61 dead blue crabs from derelict traps with completed data collection forms (ni = 380; Table 4). Additionally, volunteers indicated evidence of crabs in 24 traps by the presences of shell and carapaces. We did not detect a significant interaction effect of mesh type and Rodeo location or mesh type on estimated number of crabs in derelict traps (i.e. ^ ki ), or the probability that a trap would not contain at ^ð0Þi ) in the ZIP model. However, we detected least one crab (i.e., p

80%

60%

Stone crabs Diamondback terrapin Waterfowl and Otter

40%

Fish -Marine Fish -Freshwater

20%

0% Delacroix

Cocodrie

Pointeala Hache

Hopedale

Fig. 2. Relative composition of bycatch species recorded from derelict crab traps removed from Louisiana coastal marshes during Rodeo events in 2012 and 2013. Fish habitat categories defined in Table 2.

a significant difference in number of enmeshed crab and the probability of no crab in a trap to be significantly different across Rodeo locations (Table 4). Based on pair-wise comparisons of model coefficients, derelict traps removed during the Rodeo at Pointe a la Hache had significantly less enmeshed crab per trap than traps removed during the Rodeos at Hopedale (P < 0.001) and Delacroix (P = 0.038; Table 4). Additionally, the probability that a trap had no crab was significantly lower (P < 0.001 for all pairwise comparisons) at the Hopedale location compared to the other three locations. The probability that a trap had no enmeshed crab was similar between the Delacroix and Pointe a la Hache locations b i revealed (P = 0.898). Inspections of 95% confidence intervals on T that significantly more crab were released from derelict traps at the Delacroix location (Table 4) and likely the result of more traps removed at this location (Table 1). 4. Discussion Derelict fishing gear such as traps and nets represents an unfortunate proportion of marine debris world-wide (i.e., 10%; as re-

Table 2 Species occurrence (indicated by X) of bycatch in derelict crab traps removed from Louisiana coastal marshes in 2012–2013. Data are inclusive of all traps analyzed by citizen scientists regardless of complete data collection forms. Primary habitat of occurrence for each species was assigned according to Christmas (1973) and Ross (2001). Common Name

Blue crab Stone crab Alligator gar Spotted gar Hardhead catfish Oyster toadfish Flounder Bluegill Pumpkinseed Largemouth bass White crappie Sheepshead Silver perch Atlantic croaker Red drum Atlantic spadefish Diamondback terrapin Waterfowl unknown River otter

Scientific Name

Callinectes sapidus Menippe adina Atractosteus spatula Lepisosteus oculatus Ariopsis felis Opsanus beta Paralichthys lethosigma Lepomis macrochirus Lepomis gibbosus Micropterus salmoides Pomoxis annularis Archosargus probatocephalus Bairdiella chrysoura Micropogonias undulatus Sciaenops ocellatus Chaetodipterus faber Malaclemys terrapin Anatidae Lontra canadensis

Family

Portunidae Menippidae Lepisosteidae Lepisosteidae Ariidae Batrachoididae Paralichthyidae Centrarchidae Centrarchidae Centrarchidae Centrarchidae Sparidae Sciaenidae Sciaenidae Sciaenidae Ephippidae Emydidae Anatidae Mustelidae

Habitat

Marine Marine Freshwater Freshwater Marine Marine Marine Freshwater Freshwater Freshwater Freshwater Marine Marine Marine Marine Marine

Location Delacroix n = 724

Cocodrie n = 44

Pointe a la Hache n = 149

Hopedale n = 156

X X X X X

X X

X

X X

X X

X X X X

X X

X X

X X

X X

X X X X

X X X

X

265

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267

Table 3 Summary of traps assessed by citizen scientists for ghost fishing at Derelict Crab Trap Rodeos 2012–2013. Intactness of derelict traps was not recorded for derelict traps removed from the Hopedale closure location, and the percent of intact represents the number of traps recorded from the remaining three closures. Location

Traps (ni)

Intact (%)

Hexagonal (%)

Square (%)

Ghost fishing (%)

Delacroix Cocodrie Pointe a la Hache Hopedale

156 24 95 105

78 50 68 NA

13 42 9 69

87 58 91 31

67 33 55 88

Total

380

72

29

71

67

Table 4 Instantaneous estimates of blue crab-specific ghost fishing activity of derelict traps in coastal marshes of Louisiana 2012–2013 using model-derived Rodeo-specific estimates of ^ð0Þi ) in derelict crab traps and average number of enmeshed blue crab in actively ghost fishing derelict traps (ki), and probability of traps containing at least one crab (1  p b i were estimated from bootstrapped subsamples of traps assessed (ni) estimated total number of crab released from derelict traps. Estimates for 95% confidence bounds (CB) on T with approximated values of enmeshed crab (k0i ) and ghost fishing probability ðp0 ð0Þi Þ recalculated for each subsample. Superscripts of the same letter following estimates indicate no significant difference (P > 0.05) in pair-wise comparisons of Rodeo-specific parameters estimated within the ZIP model. Site

Traps with blue crab (gi)

Enmeshed crab (R crabti)

Estimated Enmeshed crab/ ghost fishing trap (^ ki )

Probability of trap ghost fishing ^ð0Þi ) for crab (1  p

Estimated total crab biÞ released ð T

Lower 95% bi CB on T

Upper 95% bi CB on T

Delacroix Cocodrie Pointe a la Hache Hopedale

84 6 48

275 20 128

3.13a 3.19a,b 2.43b

0.56a 0.26b 0.55a

3430.23 619.56 657.56

2600.05 60.20 481.59

4260.41 1178.93 833.53

91

333

3.55a

0.89c

1298.55

1063.43

1533.68

viewed in Good et al., 2010). The ecological and economic effects of derelict gear have been internationally realized, and removal of these underwater hazards is essential to fishery and habitat conservation. Unfortunately, removal efforts are often hampered by financial (Havens et al., 2011) or legislative (e.g., state laws; this study) limitations. However, targeting specific areas for prevention and removal of derelict fishing gear can be guided by conservation or economic objectives (Good et al., 2010; Uhrin and Schellinger, 2011). Whereas the threats of ghost fishing affect many species, targeting areas for derelict trap removal based on specific conservation concerns is still possible even if resources are limited. In our study, we found that derelict traps within the marshes around the Delacroix location likely affect a diverse community of aquatic species. Traps assessed by citizen scientists in this area contained the most number of species (Table 2), including both marine and freshwater species of fish, diamondback terrapins and mammal and bird species. Additionally, the most number of traps and estimated total number of crab enmeshed in ghost fishing derelict traps (Table 4) was observed at this location. Delacroix is located just south of the Caernarvon freshwater diversion project. Subsequently, the diversity of species found in the area below the diversions has changed to include low-salt tolerant fish and invertebrate species (Rozas et al., 2005). As such, the threat of ghost fishing may affect a more diverse set of species in marsh areas receiving large amounts of freshwater inputs. Conversely, managers and conservationists can target areas for derelict gear prevention and removal based on specific species concerns. Because reptiles cannot withstand extended periods below the water surface, they are likely to suffer mortality in derelict traps. Although not a threatened species in Louisiana, diamondback terrapins have a mortality risk associated with derelict gear (Harden and Williard, 2012; Radzio et al., 2013). Our citizen scientists recorded the greatest relative abundance of terrapins in derelict crab traps at the Cocodrie and Hopedale locations. Prevention of derelict traps in areas where terrapins are known to occur in great abundances will likely reduce mortalities associated with ghost fishing. Although we have no current estimates of blue crab abundances at any of the Rodeo locations, it is reasonable to assume that ghost fishing activity specifically related to blue crab was the greatest at

the Hopedale location where each actively ghost fishing derelict trap captured on average 3.5 crabs and the probability that a derelict trap was actively blue crab ghost fishing was nearly 90% (Table 4). Therefore, if the goal of derelict crab trap prevention or removal is to remove excess fishing pressure on a valued resource such as blue crab, areas where the fishery would best benefit can be targeted. Additionally, prior knowledge of the distribution of active, licensed commercial trap fishermen and areas of greater likelihood of crab trap loss (e.g. hurricane impacted) (Guillory, 1993) can guide managers when selecting areas for annual fishery closures and derelict gear removal efforts. The closure areas associated with the Rodeos in this study covered three parishes in Louisiana: Plaquemines, St. Bernard, and Terrebonne. Interestingly, Terrebonne parish historically has had the greatest number of licensed crab fishermen (Louisiana Department of Wildlife and Fisheries, 2011). However, the proportion of traps that were ghost fishing and the number of crab enmeshed per trap were much lower for the crab traps removed from Cocodrie, within Terrebonne parish. Crab traps (baited or derelict) can have a very high rate of escape of 40–85% (Guillory, 1993; Sturdivant and Clark, 2011). As such, blue crabs could be moving in and out of derelict traps and therefore our estimates represent instantaneous evaluations of derelict crab trap ghost fishing activity (Good et al., 2010). At all locations expect for Cocodrie, a majority (>50%) of traps were actively ghost fishing as evident by the presence of at least one enmeshed organism. Additionally, our instantaneous estimates of the average number of crabs enmeshed in derelict crab traps (2.43–3.55 crabs per trap) were slightly greater than those reported by Guillory (e.g., 0.7–1.1 crabs per trap; 1993) in simulated derelict crab traps set in Timbalier Bay during February and March. Therefore, the magnitude of our estimates compared to previous investigations suggests that ghost fishing activity may be greater than previously assumed. Citizen scientists reported a large proportion of the traps as not actively ghost fishing. There are many reasons as to why ghost fishing by derelict fishing gear occurs, and ghost fishing could be dependent on the local abundance of the organism, auto-rebaiting by dead individuals (Breen, 1987), or age of derelict gear (Akiyama et al., 2007; Good et al., 2010). The Rodeos were held during the late winter when crab catch is lowest to reduce economic hardship

266

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267

on the fishermen. Blue crab often migrate to deeper water during cooler seasons and feed less in colder water (Hines, 2007). Therefore the high proportion of empty derelict crab traps could simply represent low recruitment to inshore derelict traps during the cold winter months. Furthermore, in our design we did not have citizen scientists record physical locations of derelict traps (e.g., underwater or in marsh grass). After the Rodeos, we realized many of the retrieved derelict crab traps may have been located on the marsh edge or in the vegetation out of the water. Traps can be found in these areas due to wind and wave action, low tides, or by shrimpers who are allowed to throw traps ensnared in their nets onshore. Therefore, traps with zero enmeshed crab could have also resulted from sampling for crab where crab would not exist. Finally, not all citizen scientists indicated if a trap was still intact when it was removed, and as such a corroded or collapsed trap may not have been capable of ghost fishing. Although derelict crab trap removal efforts had not occurred in these areas in recent years there was no way of knowing how long (e.g., days or years) any of the traps were derelict. In colder waters, crab traps can continue to ghost fish for up to six years before corrosion or other effects deteriorate the gear (Maselko et al., 2013). However, warmer temperatures in Louisiana would be expected to decrease corrosion time (Maselko et al., 2013). Anecdotally, we did find that 9% (n = 14) of the traps indicated as not intact had at least one enmeshed organism. Derelict Crab Trap Rodeos were held in overlapping areas of the marshes around Delacroix and Pointe a la Hache (Fig. 1). The choice of these two closure areas by LDWF was made based on need to remove derelict crab traps from a large area of marsh during these two years, and knowledge that the overlapping areas were not well covered by volunteers in 2012. However, we separated these two locations for analyses for multiple reasons. First, the removal efforts in these locations occurred during two separate years (i.e., in 2012 at Delacroix and in 2013 at Pointe a la Hache). Hurricane Isaac made landfall on September 1, 2012 resulting in additional traps lost in this area. Second, some citizen scientists used global positioning units to record the location of removed crab traps. We evaluated these coordinates, and no overlap occurred in the location of the removed traps even though the closure areas overlapped. The difference in the ghost fishing activity supports this separation. Although number of traps assessed by citizen scientists and derelict trap ghost fishing rates (i.e., at least one enmeshed animal) were similar between these two sites, the average number of crabs per trap was significantly greater at the Delacroix location (Table 4) and the relative composition of other species recorded in traps was not similar between sites. All data were collected through a citizen science program which served the benefit of collecting a large amount of information during a short period of time. The legislation that allows for crab trap cleanup results in a tradeoff between removing many traps or collecting very detailed information on every trap removed from the water. If the state relied on staff biologists to collect data on every removed trap, a lot less traps could be removed. However, for valid data, a quality control is an important aspect of citizen scientist programs (Hidalgo-Ruz and Thiel, 2013). An anticipated quality control issue during the derelict crab trap removal Rodeos was species identification. In 2012, participants were asked to take photos of every trap. We verified all photos and all were accurate with data recorded on data collection forms. Outreach, extension, and research regarding prevention and reduction of the impacts of derelict gear should continue in Louisiana (Guillory et al., 2001a). Our study was designed to collect information regarding ghost fishing activity of derelict crab traps in Louisiana coastal marshes while engaging citizen scientists in fisheries-related research through an extension and outreach program. By utilizing citizen scientists to collect traps and data, concerned

volunteers gained better appreciation for removal and prevention of marine debris (Bonney et al., 2009). Our citizen scientists were able to witness the direct impacts of ghost fishing wherein at least 50% of derelict traps may be actively ghost fishing during cooler, winter months. While cleanup Rodeos will continue, only a small percentage of the coast can be covered in any given year. As such, continued ghost fishing-related mortalities of target species and species of concern may continue. However, interest in further development of extension programs designed to increase dialogue between fishermen, scientists, and policy makers in Louisiana and throughout the Gulf of Mexico (Guillory et al., 2001a) is promising.

Acknowledgements We thank countless volunteers, Paula Ouder, Martin Bourgeois, Carl Britt, Jon McKenzie, Bran Wagner, Paul Bangs, and anonymous reviewers. We would also like to acknowledge the National Fish and Wildlife Foundation and Louisiana Department of Wildlife and Fisheries for funding.

References Akiyama, S., Saito, E., Watanabe, T., 2007. Relationship between soak time and number of enmeshed animals in experimentally lost gill nets. Fish. Sci. 73, 881– 888. Bilkovic, D.M., Havens, K.J., Stanhope, D.M., Angstadt, K.T., 2012. Use of fully biodegradable panels to reduce derelict pot threats to marine fauna. Conserv. Biol. 26, 957–966. Bonney, R., Cooper, C.B., Dickinson, J., Kelling, S., Phillips, T., Rosenberg, K.V., Shirk, J., 2009. Citizen science: a developing tool for expanding science knowledge and scientific literacy. Bioscience 59, 977–984. Breen, P.A., 1987. Mortality of dungeness crabs caused by lost traps in the Fraser River Estuary, British Columbia. North Am. J. Fish. Manage. 7, 429–435. Christmas, J.Y., 1973. Cooperative Gulf of Mexico Estuarine Inventory and Study, Mississippi. Mississippi Marine Conservation Commission, Gulf Coast Research Laboratory, Ocean Springs, Mississippi. Coe, J.M., Rogers, D.B., 1997. Marine debris: sources, impacts, and solutions. Springer-Verlag, New York, NY. Cohn, J.P., 2008. Citizen science: can volunteers do real research? Bioscience 58, 192–197. Good, T.P., June, J.A., Etnier, M.A., Broadhurst, G., 2010. Derelict fishing nets in Puget Sound and the Northwest Straits: patterns and threats to marine fauna. Mar. Pollut. Bull. 60, 39–50. Guillory, V., 1993. Ghost fishing by blue crab traps. North Am. J. Fish. Manage. 13, 459–466. Guillory, V., McMillen-Jackson, A., Hartman, L., Perry, H., Floyd, T., Wagner, T., Graham, G., 2001a. In: Commission, G.S.M.F. (Ed.), Blue Crab Derelict Traps and Trap Removal Programs. Gulf States Marine Fisheries Commission, Ocean Springs, MS. Guillory, V., Perry, H., Van der Kooy, S.E., 2001b. The Blue Crab Fishery in the Gulf of Mexico, United States: A Regional Management Plan. Gulf States Marine Fisheries Commission, Ocean Springs, MS, p. 304. Harden, L.A., Williard, A.S., 2012. Using spatial and behavioral data to evaluate the seasonal bycatch risk of diamondback terrapins Malaclemys terrapin in crab pots. Mar. Ecol. Prog. Ser. 467, 207–217. Havens, K.J., Bilkovic, D.M., Stanhope, D., Angstadt, K., Hershner, C., 2008. The effects of derelict blue crab traps on marine organisms in the Lower York River, Virginia. North Am. J. Fish. Manage. 28, 1194–1200. Havens, K., Bilkovic, D.M., Stanhope, D., Angstadt, K., 2011. Fishery failure, unemployed commercial fishers, and lost blue crab pots: an unexpected success story. Environ. Sci. Policy 14, 445–450. Hidalgo-Ruz, V., Thiel, M., 2013. Distribution and abundance of small plastic debris on beaches in the SE Pacific (Chile): a study supported by a citizen science project. Mar. Environ. Res. 87–88, 12–18. Hines, A.H., 2007. Ecology of juvenile and adult blue crabs. In: Kennedy, V.S., Cronin, L.E. (Eds.), The Blue Crab: Callinectes sapidus. Maryland Sea Grant College, College Park, pp. 565–655. Louisiana Department of Wildlife and Fisheries, 2011. Commercial License Sales by Parish/State-License Year 2011, 6/8/2012 ed. Louisiana Department of Wildlife and Fisheries. . Louisiana Department of Wildlife and Fisheries, 2012. Recreational License Sales by Parish- License Year 2012, December 12, 2012 ed. Louisiana Department of Wildlife and Fisheries, . Louisiana Department of Wildlife and Fisheries, 2013. LDWF Monthly NonConfidential Commercial Data, Louisiana Department of Wildlife and Fisheries, 3/25/13 ed. Louisiana Department of Wildlife and Fisheries. Martin, T.G., Wintle, B.A., Rhodes, J.R., Kuhnert, P.M., Field, S.A., Low-Choy, S.J., Tyre, A.J., Possingham, H.P., 2005. Zero tolerance ecology: improving ecological

J.A. Anderson, A.B. Alford / Marine Pollution Bulletin 79 (2014) 261–267 inference by modelling the source of zero observations. Ecol. Lett. 8, 1235– 1246. Maselko, J., Bishop, G., Murphy, P., 2013. Ghost fishing in the Southeast Alaska commercial Dungeness crab fishery. North Am. J. Fish. Manage. 33, 422–431. McIlgorm, A., Campbell, H.F., Rule, M.J., 2011. The economic cost and control of marine debris damage in the Asia-Pacific region. Ocean Coast. Manage. 54, 643–651. R Development Core Team, 2013. R: A language and environment for statistical computing. R foundation for statistical computing. Radzio, T.A., Smolinsky, J.A., Roosenburg, W.M., 2013. Low use of required terrapin bycatch reduction devices in a recreational crab pot fishery. Herpetol. Conserv. Biol. 8, 222–227. Ribic, C.A., Sheavly, S.B., Klavitter, J., 2012. Baseline for beached marine debris on Sand Island, Midway Atoll. Mar. Pollut. Bull. 64, 1726–1729. Ross, S.T., 2001. Inland Fishes of Mississippi. University Press of Mississippi, Jackson, Mississippi.

267

Rozas, L.P., Minello, T.J., Munuera-Fernández, I., Fry, B., Wissel, B., 2005. Macrofaunal distributions and habitat change following winter–spring releases of freshwater into the Breton Sound estuary, Louisiana (USA). Estuar. Coast. Shelf Sci. 65, 319–336. Silvertown, J., 2009. A new dawn for citizen science. Trends Ecol. Evol. 24, 467–471. Sturdivant, S.K., Clark, K.L., 2011. An evaluation of the effects of blue crab (Callinectes sapidus) behavior on the efficacy of crab pots as a tool for estimating population abundance. Fish. Bull. 109, 48–55. Udyawer, V., Read, M.A., Hamann, M., Simpfendorfer, C.A., Heupel, M.R., 2013. First record of sea snake (Hydrophis elegans, Hydrophiinae) entrapped in marine debris. Mar. Pollut. Bull. 73, 336–338. Uhrin, A.V., Schellinger, J., 2011. Marine debris impacts to a tidal fringing-marsh in North Carolina. Mar. Pollut. Bull. 62, 2605–2610. West, J., Blanchet, H., Bourgeois, M., Powers, J., 2011. Assessment of blue crab Callinectes sapidus in Louisiana waters. Louisiana Department of Wildlife and Fisheries, Baton Rouge, p. 56.