Use a different bait type: Sea turtles

Supporting evidence from individual studies

1. A replicated, controlled study in 2002 in pelagic waters in the north-western Atlantic Ocean (Watson et al. 2005) found that unwanted catch of sea turtles was reduced when using mackerel-baited Scomber scombrus instead of squid-baited Illex spp. J-hooks, and tended to be lower when circle hooks were used in a tuna Thunnus spp. and swordfish Xiphias gladius longline fishery. Unwanted catch of sea turtles was reduced when mackerel-baited J-hooks were used (0.13–0.15 turtles/1,000 hooks) compared to squid-baited J-hooks (0.5 turtles/1,000 hooks). When mackerel-baited circle hooks were used, unwanted catch was 0.04–0.15 turtles/1,000 hooks compared to 0.05–0.21 turtles/1,000 hooks when squid-baited circle hooks were used (results were not statistically tested, see original paper for details including individual species responses). Commercially-targeted swordfish catch increased when mackerel was used (see original paper). Five hook/bait combinations were trialled: (1) 0° offset 18/0 circle hooks with 150–300 g squid bait, (2) 10° offset 18/0 circle hooks with squid bait, (3) 20°–25° offset 9/0 J-hooks with 200–500 g mackerel bait, (4) 10° offset 18/0 circle hooks with mackerel bait and (5) 20°–25° offset 9/0 J-hooks with squid bait (standard in the fishery). Thirteen vessels made 489 deployments, fishing a total of 427,382 hooks (71,000 hooks for combinations 1–4 and 142,000 hooks for combination 5). On-board observers collected catch data.

   Study and other actions tested

   Referenced paper

   Watson J.W., Epperly S.P., Shah A.K. & Foster D.G. (2005) Fishing methods to reduce sea turtle mortality associated with pelagic longlines. Canadian Journal of Fisheries and Aquatic Sciences, 62, 965-981.

2. A before-and-after study in 1994–2006 in pelagic waters off the coast of Hawaii, USA (Gilman et al. 2007) found fish-baited circle hooks reduced unwanted catch of sea turtles compared to squid-baited J-hooks in a swordfish Xiphias gladius longline fishery. Capture rates of leatherback turtles Dermochelys coriacea reduced by 83% (0.006 turtles/1,000 hooks) and loggerhead turtles Caretta caretta by 90% (0.012 turtles/1,000 hooks) when fish-baited circle hooks were used compared to squid-baited J-hooks (leatherback: 0.03 turtles/1,000 hooks, loggerhead: 0.13 turtles/1,000 hooks). Mortality rates were similar whether fish-baited circle or squid-baited J hooks were used (circle: 35 of 35 turtles survived, J: 180 of 182). Fewer turtles were deeply hooked when fish-baited circle hooks were used (leatherback: 0%, hard-shell: 22%) compared to squid-baited J hooks (10%, 60%). Swordfish catch increased by 16% after fish-baited circle hooks were introduced, but tuna (Scombridae), mahi mahi Coryphaena spp, opah Lampris spp. and wahoo Acanthocybium solandri catch reduced by 34–50% (see paper for details). Catch data from the US National Marine Fisheries Service observer programme were compared from before and after regulations were introduced requiring the use of 10° offset 18/0 circle hooks with fish bait in a pelagic swordfish longline fishery. Prior to the regulations, 9/0 J hooks with squid bait were used. ‘Before’ data used was from 1994–2002 (120 observed trips of 1,631 sets with 1,282,748 J hooks deployed) and ‘after’ data was from 2004–2006 (164 observed trips of 2,631 sets with 2,150,674 hooks deployed).

   Study and other actions tested

   Referenced paper

   Gilman E., Kobayashi D., Swenarton T., Brothers N., Dalzell P. & Kinan-Kelly I. (2007) Reducing sea turtle interactions in the Hawaii-based longline swordfish fishery. Biological Conservation, 139, 19-28.

3. A paired study in 2002–2003 in pelagic waters in the North Pacific Ocean (Yokota et al. 2009) found that using fish instead of squid as bait reduced unwanted catch of loggerhead turtles Caretta caretta in a shallow-set longline fishery. Catch of loggerhead turtles was reduced using mackerel Scomber japonicus bait (4 individuals) compared to using Japanese common squid Todarodes pacificus (18 individuals). All turtles were caught alive and subsequently released. Bait type did not alter the catch rates of commercially-targeted swordfish Xiphias gladius, bigeye tuna Thunnus obesus, blue shark Prionace glauca, and shortfin mako shark Isurus oxrinchus (see original paper for details), but increased catch of striped marlin Tetrapturus audax (mackerel: 14 individuals; squid: 5 individuals). Longlines were deployed from a single vessel (54 m long) in May-June 2002 and 2003 (19 deployments/year). Whole mackerel and squid were used as bait. The two bait species were attached alternately to standard Japanese hooks (size: 3.8-sun, 10° offset, 960 total hooks). Hooks were set to a depth of 40–90 m and lines were deployed overnight.

   Study and other actions tested

   Referenced paper

   Yokota K., Kiyota M. & Okamura H. (2009) Effect of bait species and color on sea turtle bycatch and fish catch in a pelagic longline fishery. Fisheries Research, 97, 53-58.

4. A replicated study in 2007–2008 in pelagic waters in the south of the Gulf of Garbes, Tunisia (Echwikhi et al. 2010) found that using stingray Dasyatis pastinaca as bait in a longline fishery reduced unwanted catch of loggerhead turtles Caretta caretta compared to using mackerel Scomber scombrus, but also increased catch of commercially-targeted sharks. Fewer turtles were caught with stingray bait (0.2 turtles/1000 hooks) compared to mackerel (1.2). Catch of commercially targeted sandbar sharks Carcharinus plumbeus was higher with stingray bait (19 sharks/1,000 hooks) compared to mackerel (13 sharks/1,000 hooks). J-hooks (111 mm long, 57 mm wide) were baited with pieces of stingray or whole mackerel. Data were collected by onboard observers over 21 trips on longline vessels during July–September in 2007 and 2008. In total, 48 sets of fishing gear were deployed overnight (stingray: 19, mackerel 29 deployments) using 35,950 hooks (stingray: 13,800, mackerel: 22,150 hooks) during 21 fishing trips. Fishing gear comprised a mainline (20–35 km long) with branchlines (8 m long) suspended horizontally by floats. Baited J-hooks were located at the end of branchlines approximately 40 m apart.

   Study and other actions tested

   Referenced paper

   Echwikhi K., Jribi I., Bradai M.N. & Bouain A. (2010) Effect of type of bait on pelagic longline fishery–loggerhead turtle interactions in the Gulf of Gabes (Tunisia). Aquatic Conservation: Marine and Freshwater Ecosystems, 20, 525-530.

5. A controlled study in 2001–2010 in sea water test tanks in Italy (Piovano et al. 2012) found that loggerhead turtles Caretta caretta were less likely to bite at mackerel Scomber scomber bait than squid Ilex argentinus bait, and that this varied with the size of the turtle. Overall, turtles were less likely to bite at mackerel bait (biting at mackerel: 23% frequency) than squid (biting at squid: 60% frequency), but more likely to bite at mackerel than no bait (5–8% frequency). Smaller turtles were more likely to bite at mackerel bait and larger turtles were more likely to bite at squid bait (data reported as statistical model outputs). Whole mackerel and squid were selected as bait as these are commonly used in longline fisheries. Individual turtles (30 in total) were presented with bait of the same species (13 mackerel tests; 20 squid tests) and no bait in three different coloured sacks and their response was recorded on a portable video camera. Three turtles were tested using both bait types. Attempts to bite a sack were considered proof of biting behaviour. Turtles were wild caught individuals who had been in the rescue centre for <4 months and were considered fit to be released.

   Study and other actions tested

   Referenced paper

   Piovano S., Farcomeni A. & Giacoma C. (2012) Effects of chemicals from longline baits on the biting behaviour of loggerhead sea turtles. African Journal of Marine Science, 34, 283-287.

6. A replicated, controlled study in 2008–2012 in pelagic waters in the Southern Atlantic (Santos et al. 2013) found that using mackerel Scomber spp. bait reduced unwanted catch of loggerhead turtles Caretta caretta and leatherback turtles Dermochelys coriacea compared to using squid Illex spp., but when squid bait was used, unwanted catch rates depended on the hook type used. Unwanted turtle catch was lower when mackerel bait was used regardless of hook type (non-offset circle: 0.2 turtles/1,000 hooks; offset circle: 0.2; J-hook: 0.3), compared to squid bait. The number of turtles caught by squid bait varied with hook type (non-offset circle hooks: 0.7 turtles/1000 hooks; offset circle hooks: 0.6 turtles; J-hooks 1.7). This pattern was observed for both leatherback and loggerhead turtle species (see paper for details). Overall turtle mortality rates were similar regardless of whether squid (146/228 individuals alive) or mackerel (40/58 individuals alive) was used as bait. Three hook types baited with either squid or mackerel were used alternately on a commercial longline fishing vessel: traditional J-hook (size: 9/0) and two circle hooks (a non-offset and a 10ᵒ offset, both sized: 17/0; 148,800 total hooks/type). In total 310 longline deployments (1,440 hooks/deployment; 446,400 total hooks, lines set to 20–50 m depths) were carried out overnight in October 2008–February 2012. One bait type was used in each deployment. Turtle catch was monitored by onboard observers.

   Study and other actions tested

   Referenced paper

   Santos M.N., Coelho R., Fernandez-Carvalho J. & Amorim S. (2013) Effects of 17/0 circle hooks and bait on sea turtles bycatch in a Southern Atlantic swordfish longline fishery. Aquatic Conservation: Marine and Freshwater Ecosystems, 23, 732-744.

7. A replicated, paired, controlled study in 2008–2011 in pelagic waters in the north-east Atlantic Ocean (Coelho et al. 2015) found that changing from squid Illex spp. bait to mackerel Scomber spp. bait reduced unwanted catch of hard-shell sea turtles (Cheloniidae spp.), but not leatherback turtles Dermochelys coriacea in a pelagic longline swordfish Xiphias gladius fishery. Unwanted catch of hard-shell sea turtles (loggerhead Caretta caretta, olive ridley Lepidochelys olivacea and Kemp’s ridley Lepidochelys kempii turtles) was reduced when mackerel bait was used (0.07–0.16 turtles/1,000 hooks) compared to squid (0.14–0.35 turtles/1,000 hooks). Unwanted catch of leatherback turtles was similar when mackerel (0.39–0.95 turtles/1,000 hooks) or squid (0.50–0.10 turtles/1,000 hooks) bait was used. In August 2008–December 2011, a commercial vessel carried out 202 overnight longline fishing deployments (lines: 55 nm long with 5 branchlines, deployed 20–50 m deep, lit by green lights). Whole squid or mackerel were used as bait (one type of bait/line deployment). Hook styles (10° offset J-hooks traditionally used in the fishery; non-offset G-style circle hooks; and 10° offset Gt-style circle hooks) were alternated every 70–80 hooks along the line in a randomized start order (254,520 total hooks deployed with 42,420 of each hook/bait combination). Unwanted catch was counted and released.

   Study and other actions tested

   Referenced paper

   Coelho R., Santos M.N., Fernandez-Carvalho J. & Amorim S. (2015) Effects of hook and bait in a tropical northeast Atlantic pelagic longline fishery: Part I-Incidental sea turtle bycatch. Fisheries Research, 164, 302-311.

8. A replicated study in 2004–2011 in pelagic waters in the Eastern Pacific Ocean (Parga et al. 2015) found that using fish instead of squid bait reduced the likelihood of olive ridley turtles Lepidochelys olivacea swallowing hooks in an artisanal surface longline fishery. When fish bait was used, hooks were less likely to be swallowed by olive ridley turtles than when squid bait was used (results reported as odds ratios, see original paper for details). Using fish bait in combination with larger circle hooks lead to the largest proportion of external hookings. In 2004–2011, incidental catch rates of sea turtles on circle hooks (sizes 12/0–18/0), tuna hooks and traditional J-hooks (see original paper for hook details) were compared by placing hooks in alternative sequence along longlines (3.5 million total hooks used in 8,996 line deployments) in an artisanal longline fishery. Bait used was classed as squid (Dosidicus gigas, Illex sp. and Loligo sp.) or fish (Opisthonema spp., Scomber japonicus, Auxis spp. and Sardinops sagax) and only deployments using one type of bait were included in the analysis (4,838 of 8,996 line deployments). Information on hooking location and entanglement of sea turtles was recorded (1,823 total olive ridley turtles).

   Study and other actions tested

   Referenced paper

   Parga M.L., Pons M., Andraka S., Rendón L., Mituhasi T., Hall M., Pacheco L., Segura A., Osmond M. & Vogel N. (2015) Hooking locations in sea turtles incidentally captured by artisanal longline fisheries in the eastern Pacific Ocean. Fisheries Research, 164, 231-237.

9. A replicated, before-and-after study in pelagic longline fisheries in the Atlantic and North Pacific (Swimmer et al. 2017) found that using fish bait resulted in less leatherback Dermochelys coriacea and loggerhead Caretta caretta turtle bycatch compared to when squid bait was used. The number of turtles caught on longlines was lower in the Atlantic when fish bait was used (leatherback: 0–6% chance with circle hooks, 13% with J hooks; Loggerhead: 0–5% with circle hooks, 9% with J hooks) compared to squid bait (leatherback: 9% with circle hooks, 20% with J hooks; Loggerhead: 11% with circle hooks, 18% with J hooks). The same was true in the Pacific (loggerhead: circle hook: 1% with fish, 2% with squid; j hook: 5% with fish, 13% with squid). Following the introduction of regulations on bait and hooks, overall turtle bycatch was reduced in both the Atlantic (leatherback: 40% reduction; loggerhead: 61% reduction) and Pacific (leatherback: 84% reduction; loggerhead 95% reduction). Fisheries were closed in 2001 and re-opened with regulations regarding bait (fish or squid) and hook type (circle or J hooks) (see paper for details). Pelagic Observer Program data from before (1992–2001) and after (2004–2015) regulations was used to determine the number of turtles caught/1,000 hooks.

   Study and other actions tested

   Referenced paper

   Swimmer Y., Gutierrez A., Bigelow K., Barceló C., Schroeder B., Keene K., Shattenkirk K. & Foster D.G. (2017) Sea turtle bycatch mitigation in U.S. longline fisheries. Frontiers in Marine Science, 4, 260.