Short Stories

6 pages
3 views

Food color and marine turtle feeding behavior: Can blue bait reduce turtle bycatch in commercial fisheries

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Food color and marine turtle feeding behavior: Can blue bait reduce turtle bycatch in commercial fisheries
Transcript
  MARINE ECOLOGY PROGRESS SERIESMar Ecol Prog SerVol. 295: 273–278, 2005Published June 23 INTRODUCTION The incidental capture of marine turtles in longlinefishing gear has been reported to be a significant fac-tor contributing to the current decline of sea turtle pop-ulations (Hall et al . 2000, Lewison et al . 2004). Sea tur-tles most commonly interact with shallow-set (<100 m)fishing gear that targets swordfish Xiphias gladius  ,mahi mahi, or dolphin Coryphaena hippurus  , or sur-face-feeding tunas (Ferreira et al. 2001, Polovina et al.2003). Loggerhead turtles Caretta caretta and oliveridley Lepidochelys olivacea turtles generally bitebaited hooks, whereas leatherback turtles Dermo-chelys coriacea are most often hooked in the flippers orbecome entangled in the fishing lines (Witzell 1999).Despite the fact that both species apparently forageprimarily near or at the surface, loggerhead turtles areopportunistic feeders, with a broad range of prey items(Parker et al. 2005), whereas leatherback turtles morespecifically target jellyfish (James & Herman 2001).Methods to reduce incidental capture of turtles inlongline fisheries include area and seasonal closures.While such closures can reduce sea turtle interactionrates, they are not likely to be adopted internationallybecause of the potential for severe economic disrup-tion (Brothers et al. 1999). Moreover, establishing and © Inter-Research 2005 · www.int-res.com*Email: yonat.swimmer@noaa.gov Food color and marine turtle feeding behavior: Can blue bait reduce turtle bycatch in commercial fisheries? Yonat Swimmer 1, *, Randall Arauz 2 , Ben Higgins 3 , Lianne M c Naughton 4 , Marti McCracken 1 , Jorge Ballestero 2 , Richard Brill 5 1 National Marine Fisheries Service, Pacific Island Fisheries Science Center, 2570 Dole Street, Honolulu, Hawaii 96822, USA 2 PRETOMA & Sea Turtle Restoration Program, 1203-1100 Tibás, San José, Costa Rica 3 National Marine Fisheries Service, Galveston Sea Turtle Facility, 4700 Avenue U, Galveston, Texas 77551, USA 4 Joint Institute for Marine and Atmospheric Research, University of Hawaii, 1000 Pope Road, Honolulu, Hawaii 96822, USA 5 Virginia Institute of Marine Science & National Marine Fisheries Service, 1208 Greate Road, Gloucester Point, Virginia 23062, USA ABSTRACT: We conducted laboratory and field experiments to investigate the behavioral responsesof Kemp’s ridley Lepidochelys kempii  and loggerhead turtles Caretta caretta to whole squid dyeddifferent colors. Our ultimate goal was to identify bait modifications that could reduce the interactionof turtles with longline fishing gear. In captivity, both turtle species clearly preferred untreated squidover squid that had been dyed dark blue. Loggerhead turtles also preferred untreated squid over red-dyed squid, whereas Kemp’s ridley showed the opposite response. Field trials of blue-dyed bait wereconducted on commercial fishing boats in the Gulf of Papagayo, Costa Rica, where the incidentalcapture of olive ridley turtles Lepidochelys olivacea is high (long-term average, approximately 7turtles per 1000 hooks). We found no differences in rates of turtle interactions (8.4 and 8.1 individu-als per 1000 hooks) when using untreated versus blue-dyed baits. Although effective in laboratorysettings with captive turtles, dyeing bait does not appear to have potential as an effective mitigationmeasure to reduce sea turtle bycatch in longline fisheries.KEY WORDS: Sea turtle fisheries interactions· Bait color· Sea turtle bycatch mitigation Resale or republication not permitted without written consent of the publisher   Mar Ecol Prog Ser 295: 273–278, 2005 enforcing marine protected areas on the high seas isnot a priority and thus not likely to be adopted by anycountry with fishing interests. What is urgently neededis a readily exportable solution to the worldwideproblem of sea turtles incidentally being captured inlongline fishing gear, such as simple gear or baitmodification.A number of operational modifications (e.g. use ofstreamer lines, underwater line shooters, weights toenhance hook sinking rates, night setting, side settingand use of blue-dyed bait) have been shown to reduceseabird bycatch in longline fisheries (Brothers et al . 1999), while simultaneously maintaining viable fishingoperations. Subsequently, numerous fishery manage-ment regimes worldwide have adopted seabird mitiga-tion measures. These results served as a model for ourstudies.Herein, we present work aimed at identifying a sim-ple mitigation method that could reduce sea turtleinteractions with longline fishing gear based on thevisual appearance of bait. We chose to investigate theeffects of bait color on turtles’ feeding behavior,because squid bait dyed dark blue has been shown tosignificantly reduce seabird interactions in longlinefisheries, while simultaneously maintaining acceptablyhigh catch rates of the targeted fish species (McNa-mara et al . 1999). We also aimed to understand the sen-sory modalities that sea turtles use to find food sourcesthat attract them to fishing gear and that influencetheir ‘bite/no bite’ decision. Specifically, we seek tounderstand marine turtles’ reliance upon olfaction andvision in their food search and feeding behaviors. MATERIALS AND METHODSCaptive experiments. We conducted experimentswith 49 two yr old loggerhead Caretta caretta and42two yr old Kemp’s ridley Lepidochelys kempii  turtlesat the National Marine Fisheries Service Sea TurtleFacility (STF) in Galveston, Texas. The average massand straight carapace length (SCL) of loggerhead tur-tles were 11.14 kg and 45.0 cm, respectively, and thoseof Kemp’s ridley turtles were 5.70 kg and 33.26cm,respectively. The loggerhead turtles hatched fromapproximately 3 to 6 nests on the beach in Clearwater,Florida, on 25 August 1999, and were collected andcaptive-reared under permits issued by the FloridaFishand Wildlife Conservation Commission (TPNo.015)and the U.S. Fish & Wildlife Service (No. TE676379-2).The Kemp’s ridleys hatched from approximately 40nests at Rancho Nuevo, Mexico, on 17 August 2000,and were part of a donation from the Mexican Govern-ment to the United States. The Kemp’s ridley hatchlingswere collected and captive reared under U.S. CITESpermit No. US9242929, CITES export permit MEX-10941, USFWS TE676379-2, DAN-02768 and Permisode Pesca de Fomento No. 25.05.00.213.03-1320.Details of animal husbandry at the STF are describedby Higgins (2003). Briefly, turtles were housed individ-ually in suspended cages within large common seawa-ter tanks measuring 6 m long × 2 m wide × 60 cm deep.The cages holding the loggerheads and Kemp’s ridleysmeasured 90 × 168 × 46 cm deep (rectangular) and76cm diameter × 46 cm deep (round), respectively. Airtemperature was regulated at 30°C, water temperaturewas between 25 and 30°C and salinity was between14and 32 ppt. Turtles were fed Purina Aquamax500(Purina Mills) grower pellet food at 1 to 3% bodyweight d –1 , which resulted in maximum growth with-out overfeeding (Higgins 2003). Prior to behavioraltrials, turtles were fasted for approximately 48 h.To prepare colored baits, frozen squid (Sea WaveCalamari, Monterey Fish Co.) were thawed and mari-nated overnight in either a 1% solution of blue-color-ing (Food Dye F, D and C No.1) or a 1% solution ofred-coloring (Food Dye F, D and C No. 40) dye (Vir-ginia Dare).Behavioral trials were conducted on individual ani-mals in outdoor tanks (4 m diameter and 1.5 m deep).Tank sides and bottom had been previously paintedlight blue. Turtles were moved into the tanks and al-lowed to acclimate for 30 min. Then, 2 dyed and 2un-treated squids (‘baits’) were presented simultaneously.Baits were suspended from a horizontal plastic rod andwere spaced approximately 10 cm apart. The order ofthe 4 baits across the rod was randomized by tossing acoin. The baits were submerged, and the order inwhich baits were eaten was recorded. Trials lasted30min after which the turtles were returned to theirholding tank. If a turtle did not make a food choicewithin this time, it was excluded from the analysis. A‘trial’ is herein defined as an individual experimentwith 1 turtle and 1 ‘treatment’ (e.g. blue- or red-dyedsquid).Individual loggerhead turtles were used 4 timesbetween October 2001 and March 2002 to test for pref-erences between untreated squid and blue-dyed squid(n = 49). The same individuals were used once duringMarch 2002 to test for preferences between untreatedand red-dyed squid (n = 42). Experiments with Kemp’sridley turtles (n = 28) were conducted during July andAugust 2002. There was a minimum of 2 wk betweenfood preference trials.We investigated turtles’ preferences by testing thenull hypothesis that turtles would select (as their firstchoice) both treated (e.g. red- or blue-dyed) squidwith the same frequency (50%) as untreated squid.Bait selected as the first choice was used to definepreference. Binomial tests were conducted for each 274  Swimmer et al.: Food color and marine turtles dyed food trial using S-Plus V. 6.2 for Windows(Insightful). Field trials. Experiments to test the efficacy of dyedbait in reducing sea turtle bycatch were conducted on2 commercial longline fishing boats (length: ~15 m),the ‘Don Miguel’ and the ‘Don Roberto’, operating outof Playas del Coco in the Gulf of Papagayo, Costa Rica.Trials on both boats were conducted simultaneouslyfrom 1 to 15 December 2003. The boats fished as closeto each other as feasible in the same general areabetween 7°00’N, 86°00’W and 10°30’N, 88°00’W(Fig.1) and targeted mahi mahi Coryphaena hippurus  and tunas Thunnus spp.. Lines were deployed in themorning and remained in the water for approximately8 h before haulback in the afternoon. Whole squid Loligo  spp. was used exclusively as bait for the first 6sets on both boats. During sets 7 through 11, musclestrips cut from sailfish Istiophorus platypterus werealso used as bait on both boats. Baits were soaked inthe same 1% solution of blue food coloring used in thelaboratory trials. The boats made 22 longline sets, 9with blue-dyed bait and 13 with untreated bait. Setswere randomly alternated between blue-dyed anduntreated bait. Size 12/0 circle hooks were used exclu-sively. The average number of hooks per set was 560(min. = 452, max. = 675, SD = 82.6) for the ‘Don Miguel’and 606 for the ‘Don Roberto’, which included 1 trip inwhich only 250 hooks were set due to an accidentonboard (min. = 250, max. = 675, SD = 124.8). Hookswere relatively shallow given that float and branchlines were approximately 3 to 4 fathoms. Catch rateswere standardized to 1000 hooks.Because of small sample sizes of squid-only baitsthat were successfully dyed deep blue, only qualitativeanalyses of the data were performed. RESULTSCaptive turtle trials Loggerhead turtles Caretta caretta displayed a clearpreference for untreated squid when presented with achoice between blue and untreated squid. A total of98% of the turtles selected untreated squid over blue-dyed squid during each of 4 presentations, which wassignificantly different from the expected value of 50%(p < 0.0001; Fig. 2). Loggerhead turtles also preferreduntreated bait over red-dyed bait (p = 0.0007; Fig. 3)during their 1 experimental trial. Kemp’s ridley turtles Lepidochelys kempii  selected untreated squid overblue-dyed squid at a frequency greater than expected,both for their first and second trials (p = 0.0225 and p =0.0007, respectively; Fig. 4). In contrast to loggerheadturtles, they selected red-dyed squid over untreatedsquid with statistically greater frequency during bothfood trials (p < 0.02; Fig.5). Field trials During both trips, 108 olive ridley turtles Lepido-chelys olivacea and 7 green turtles Chelonia mydas  275Fig. 1. Location of fishing activities for commercial fishing boats, the ‘Don Miguel’ ( Q ) and the ‘Don Roberto’ ( s ) 05101520253035404550Trial 1Trial 2Trial 3Trial 4    N  u  m   b  e  r  o   f   t  u  r   t   l  e  s Fig. 2. Caretta caretta . Loggerhead turtles’ first choice be-tween untreated (open bars) and blue-dyed squid (black bars)(n = 49 in all 4 trials)Fig. 3. Caretta caretta . Loggerhead turtles’ first choice be-tween untreated and red-dyed squid (n = 42)  were caught (Table 1). Because the number of greenturtles was relatively small, data on turtle catch ratesare combined to include both green and olive ridleyturtles. Turtle catch rates were similar for sets madewith both blue and untreated bait (8.4 and 8.1 individ-uals per 1000 hooks, respectively (Table 2). DISCUSSION Our idea was to develop a bait modification thatwould reduce the incidence of hooking threatened andendangered sea turtles to acceptable levels, but onethat would maintain an economically viable catch rateof the targeted species.As an ideal, we also aimed toidentify a mitigation measure with demonstrated abil-ity to offer an economic advantage to fishers, therebyincreasing its potential use in an unregulated fishery.This could be achieved most readily by decreasing baitloss to unwanted bycatch species and increasingbaited hooks available for commercial fish.Our results clearly show that captive loggerhead andKemp’s ridley turtles base food choice and ‘bite/nobite’ decisions on color alone and that both species 024681012141618Trial 1Trial 2    N  u  m   b  e  r  o   f   t  u  r   t   l  e  s Fig. 4. Lepidochelys kempii  . Kemp’s ridley turtles’ first choicebetween untreated (open bars) and blue-dyed squid (blackbars) (n = 42 in both trials) 024681012141618Trial 1Trial 2    N  u  m   b  e  r  o   f   t  u  r   t   l  e  s Fig. 5. Lepidochelys kempii  . Kemp’s ridley turtles’ first choicebetween untreated (open bars) and red-dyed squid (blackbars) (n = 42 in both trials)Mar Ecol Prog Ser 295: 273–278, 2005276‘Don Miguel’ ‘Don Roberto’Total%CPUETotal%CPUEHooks trip –1 61656669 Sharks Black tip 52.10.8100.000.00 Carcharhinus limbatus  Silky 187.62.923814.965.70 C. falciformis  Oceanic whitetip 20.80.3200.000.00 C. longimanus  Bigeye thresher 41.70.6510.390.15 Alopias superciliosus  Other fish Mahi mahi 7531.512.173513.785.25 Coryphaena hippurus  Yellowfin tuna 198.03.08155.912.25 Thunnus albacares  Sail fish 166.72.60197.482.85 Istiophorus platypterus  Skipjack tuna 10.40.16114.331.65 Katswonus pelamis  Black marlin 62.50.9741.570.60 Makaira indica Blue marlin 31.30.4951.970.75 M. mazara Striped marlin 10.40.1672.761.05 Tetrapterus audax  Dragon fish 20.80.3200.000.00 Gempylus serpens  Flying fish 00.00.0010.390.15Family AtherinidaeManta ray00.00.0020.30Sting rays3414.35.525320.877.95 Sea turtles Olive ridley 4719.77.626124.029.15 Lepidochelys olivacea Pacific green 52.10.8120.790.30 Chelonia mydas agassizi  Table 1. Total catch statistics for fishing vessels ‘Don Miguel’and ‘Don Roberto’ after 11 sets per boat (CPUE: catch per unit effort)Bait type‘Don Miguel’‘Don Roberto’CombinedaverageUntreated7.98.98.4(SE = 3.13; n = 6)(SE = 3.14; n = 7)(n = 13)Blue dye8.97.28.1(SE = 3.43; n = 5)(SE = 3.55; n = 4)(n = 9)Table 2. Lepidochelys olivacea and Chelonia mydas. Averagenumber of olive ridley and green sea turtles, combined,caught per 1000 longline fishing hooks by the 2 commercialfishing boats after 22 sets in the eastern tropical Pacific Ocean  Swimmer et al.: Food color and marine turtles preferred untreated squid over blue-dyed squid. Previ-ous studies on the responses of captive green turtles Chelonia mydas  to foods of various colors found thatblue-dyed squids were also significantly less preferredthan untreated squid (Swimmer & Mailloux 2003).Because of our findings that 3 species of sea turtleshave an aversion to blue-dyed squid, we designed fieldtrials employing commercial longliners to see if thebehaviors of captive turtles could be translated toreduced levels of turtle bycatch. Surprisingly, use ofblue-dyed bait in these trials did not result in reducedturtle bycatch.Dying bait blue can increase catch rates of fish inpelagic fisheries, and the procedure is employed rou-tinely in longline fisheries, despite the fact that no sta-tistical analyses have been conducted (McNamara etal. 1999). Numerous studies have also found that use ofblue-dyed bait successfully reduces seabird bycatch inlongline fisheries (McNamara et al. 1999), for somespecies (e.g. albatross species) by as much as 95%,presumably because the dyed bait is more difficult forbirds to detect as it reduces the contrast between thebait color and the wavelength of upwelling light. In ourstudies with captive turtles, however, the dark-blue-dyed squids were clearly visible against the light bluewalls of the holding tank.Both behavioral and physiological studies of colorvision in various sea turtle species and age classes con-firm that they can detect light in the 400 to 700 nmrange (i.e. from blue to red; Fehring 1972, Dvorak &Granda 1990, Bartol & Musick 2003). These findingssuggest that turtles in our experiments could distin-guish blue-dyed squid from untreated squid and thatthey had an aversion to biting blue-dyed squid. Pre-vious studies also suggest that the sea turtles’ ability todistinguish colors is an important component in theirforaging ecology (Fehring 1972). It is difficult to inter-pret the visual differences between the captive turtles’environment within a tank as compared to the percep-tion of bait in the open ocean. In addition to the angleat which the animal looks at the bait, the visibility ofobjects and colors differs depending on the back-ground light. Thus, an animal looking up at bait to-wards a bright light would have a different perceptionof bait color than a turtle looking downward into thedark water. Factors that affect aquatic vision are com-plex and include the organisms’ own visual capabili-ties, the depth and angle of the viewed object, as wellas the optical properties of the water (Johnsen 2002). Inthe case of sea turtles, recent evidence suggests thatyoung loggerhead turtles can see ultraviolet (UV) lightand that they have multiple combinations ofcolor re-ceptors (K. Fritsches pers. comm.), thus enabling themto differentiate shades of colors that appear the same tohumans. However, for the open ocean and variousdepths, the exact light conditions and wavelength dis-crimination abilities of sea turtles are currently un-known, and thus we assume that objects in turtle tanksdo not appear the same as in the open ocean.Captive loggerhead turtles fed a particular diet havebeen shown to prefer that diet over time (Grassman &Owens 1982). Additionally, behavioral studies withgreen turtles also found that turtles’ remembered spe-cific food items (Angermeier & Hidalgo 1996). In ourstudy, however, the responses of loggerhead andKemp’s ridley turtles to foods of various colors werelikely innate responses, as these turtles had beenraised since hatching on an artificial diet. Similar pref-erences for srcinal food items were found for greenturtles that had been brought into captivity as large juveniles from their inshore foraging grounds (Swim-mer & Mailloux 2003). These turtles clearly preferredfood items that were eaten first while in captivity.Additionally, these turtles (n = 23) preferred untreatedsquid over blue-dyed squid.During the field trials on commercial longline fishingboats within the Exclusive Economic Zone of CostaRica, blue-dyed bait did not reduce catch rates of oliveridley or green sea turtles. This was surprising giventhat longline gear in Costa Rica was set very shallow(<10 m) and that this visual environment most closelymatched the environment of experiments with thecaptive turtles. Use of blue-dyed squid bait was alsofound to be ineffective at reducing rates of sea turtlebycatch in the North Atlantic Ocean during field trialsconducted on commercial longline fishing vesselsover2 fishing seasons (J. Watson et al. unpubl. data;available at http://www.mslabs.noaa.gov/mslabs/docs/pubs.html). Additional experiments were conducted todetermine the catch rates of targeted swordfish usingfishing gear designed to be less detectable to turtles inthe North Pacific Ocean. In order to achieve ‘stealth’gear, floats were counter shaded, lines were blue,hardware (e.g. snaps) was dulled, LED light sticksfaced downward and blue-dyed squid baits were usedexclusively. As a result, however, significantly fewer(30% reduction) swordfish were caught as comparedto controls. This study was not designed to detect anyeffect on sea turtle captures because the expected (andobserved) number of turtles captured was too few(Boggs 2003).Our results with marine turtles in captivity demon-strate the importance of color in the turtles’ decision tobite bait. Additionally use of blue bait has been aneffective mitigation measure to reduce seabird by-catch. Results presented herein and in other studies,however, suggest that dying bait blue is not effective inreducing sea turtle bycatch. Thus, we do not believethat further investigations regarding use of blue bait toreduce sea turtle bycatch are warranted. Because less- 277
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x