Fisheries bycatch is considered one of the main threats to sea turtles. Previous studies in Tunisia recognized ray nets named Garrasia as one of the fishing gears with the highest levels of by-catch. In this study, 56 onboard observations were conducted with fishermen in the south of Tunisian coastal in ports of Zarzis and Jerba between 2018 and 2022. Eighty-three sea turtles were incidentally caught in fishing sets. All sea turtle captured were loggerhead Caretta caretta and 93.94% of individuals were dead. The majority of individuals caught were classified as sub-adults and adults (80.72%), with an average carapace length (CCL) of 67.33± 2.6 cm.

The mean catch per unit effort (CPUE) is high: 0.46 turtles * net km^-1 * 24 h^-1; 95% CI: [0.27-0.65] for Ray nets with seasonal variations. The highest CPUE being observed in spring (mean CPUE 0.62 ± 0.65 turtles * net km^-1 * 24 h^-1) followed by autumn (CPUE = 0.46± 0.23 km-1 24 h-1) and summer (CPUE= 0.38± 0.24 km^-1 24 h^-1).

Based on available census of active ray fisheries in Tunisia, the total capture by ray nets is estimated 1944.63 turtles per year (95%CI [1118.04 - 2691.58617]).

The important interaction of sea turtles with Garrasia objectified in our study confirms that the gulf of Gabes is an important foraging area and developmental habitat for loggerhead turtles where fisheries interaction with sea turtle is high.

Results emphasize the need of identification and implementation of mitigation measures to reduce sea turtle bycatch by small fisheries with particular attention for large meshes gillnets.

Keywords: sea turtles, raies, CPUE, gillnet, bycatch, Gulf of Gabes, Tunisia, Mediterranean sea.

Fisheries bycatch is the greatest threat to sea turtle populations worldwide. Recent work has shown that the use of artificial green light on fishing nets has considerable potential to reduce sea turtle bycatch. To understand how green lights lead to reduced bycatch of sea turtles, we conducted experiments in a controlled setting to detail the effect of green lights on entanglement behavior. We recorded interactions of wild-caught juvenile green and loggerhead turtles within experimental treatments including modified gillnets and/or green LED lights in a test tank at the St. Lucie Nuclear Power Plant in Jensen Beach, Florida. Our nighttime binary choice trials consisted of a green-LED illuminated gillnet vs. a control (unlighted) gillnet, an illuminated gillnet vs. a green LED light, and an activated green LED light vs. a deactivated green LED light. Netted treatments included a pathway with a gillnet fully blocking it, and netless treatments included an open pathway, either with an activated or deactivated green LED light. We compared binary choices made by turtles between treatments and measured turtles’ avoidance and escape behaviors. Our data showed that loggerheads tend to become entangled sooner and more frequently than green turtles. Behavioral observations and treatment choices indicated that both species were attracted to green LED lights. Illuminated nets did not reduce entanglement as much as field studies have reported. This could be due to mixed effects from lighting, including a light trapping effect. Further testing is necessary to determine whether turtles are more likely to become entangled directly under the light due to a light trapping effect or if a larger net creates a different stimulus than our smaller modified gillnets.

Pound nets are a passive fishing gear used globally that can lead to frequent bycatch of sea turtles and other protected marine megafauna. However, despite their global ubiquity and high bycatch, there are few tested solutions to reduce sea turtle bycatch in pound net fisheries. Net illumination is a sensory-based bycatch reduction technology that has reduced bycatch of sea turtles and other marine megafauna while maintaining target fish catch in multiple coastal gillnet fisheries worldwide. Given the success of gillnet illumination, we tested the effects of illuminating pound nets with green light-emitting diodes (LEDs) on sea turtle bycatch and target fish catch in a coastal North Carolina pound net fishery. We conducted 133 pound net trials (N = 72 control nets; 61 illuminated nets) across 3,153 hours of fishing effort, resulting in 93 sea turtles captured (N = 58 turtles in control nets; 35 turtles in illuminated nets). We fit generalized additive models (GAMs) to quantify the effects of illumination and environmental variables on sea turtle and fish catch rates. The best-fit GAM for sea turtles predicted a significant 28% decrease in bycatch rates (defined as the number of turtles caught per 24 hours of fishing effort) in illuminated versus control nets. Large turtles (> 40 cm, comprised of 94% loggerheads) were predicted to have a much greater reduction in bycatch rates (-46%) in illuminated nets compared to small turtles (-4%, < 40 cm, comprised of 65% green turtles and 35% Kemp’s ridley turtles). Additionally, the predicted reduction in the bycatch of recaptured turtles (-52%) was greater than the reduction in bycatch for newly captured turtles (-15%). Best-fit GAMs also predicted a significant 64% reduction in shark and ray bycatch rates. There was no significant reduction in target flounder catch, although the trials were conducted outside of the typical fishing season and with different mesh size than is typically used in the fishery. Moreover, illuminated pound nets significantly reduced bycatch rates of sublegal flounder, which may have important implications for recovery of the stock. Our results suggest that pound net illumination using green LEDs may be an effective method for reducing sea turtle and elasmobranch bycatch. While this study was conducted using surface set pound nets, our results provide a global framework for submerged pound net fisheries with high sea turtle mortality.

Bycatch, the incidental capture of non-target species, is among the greatest threats to sea turtle populations worldwide. Numerous studies have found sea turtle bycatch to be highest in small-scale fisheries (SSF). To reduce these rates, bycatch reduction technologies (BRTs) have been developed to improve the sustainability of SSF and conserve vulnerable marine species, especially sea turtles. We compiled peer-reviewed studies to determine how BRTs are assessed in small-scale passive fisheries of developing nations worldwide for sea turtles and other air-breathing marine megafauna. We assessed 34 experiments from 24 published studies that tested seven BRTs in gillnets (n = 25), longlines (n = 8), and a trammel net (n = 1). Most studies report the use of net illumination (n = 13), acoustic deterrents (n = 8), and circle hooks (n = 7). These technologies either reduced or had no impact on bycatch rates while maintaining or increasing target catch rates. Few BRTs experiments assessed target catch market value. Over half of the BRTs were for sea turtles (61.8%), followed by cetaceans (41.2%), and seabirds (5.9%). Bycatch declined 85.7% of the time for sea turtles, 57.1% for cetaceans, and 100% for seabirds. Additionally, 11 experiments (32.4%) included at least one quantified assessment of fisher metrics and eight experiments (23.5%) included at least one anecdotal assessment of fisher metrics, such as BRT expense, change in operational efficiency, change in ease of use, effects on fishing gear, effects on fisher safety, or other benefits and costs. Our findings highlight the need for researchers testing BRTs to include a broader range of socio-economic and fishery efficiency metrics to better assess and improve the adoption potential of these tools for sea turtles and other marine megafauna worldwide.

Bycatch is globally recognised as a major threat to marine megafauna, with more recent recognition of the role played by small scale fisheries (SSFs). The Mediterranean fishing fleet primarily consists of SSFs and the sea is considered one of the most overexploited worldwide. Concerns regarding impacts of SSFs in the eastern basin, particularly that of marine turtle bycatch, have been shared for several decades, but to date relatively few detailed studies have addressed this priority research area. Here, we used a combination of strandings data, onboard observations, fisher self-reporting and opportunistic sampling over a four-year period to quantify the magnitude and identify key drivers of marine turtle bycatch within the Northern Cyprus SSF fleet. Across all sampling methodologies, 796 marine turtles were recorded as stranded (n=623) or incidentally captured (n=173) including 411 green turtles (Chelonia mydas), 364 loggerhead turtles (Caretta caretta), and one leatherback turtle (Dermochelys coriacea). Most green turtles were of juvenile size (mean±SD=40.3±16.7 cm), whereas loggerhead turtles were mostly subadult and adult size (mean±SD=64.7±12.0 cm). Green turtle incidental capture rates were highest in trammel nets and shallower set fishing operations, whereas, loggerhead turtle capture rates were highest in gill nets and deeper set fishing operations. Bycatch probability in set nets significantly differed between green and loggerhead turtles and was largely driven by the effect of individual fisher behaviour, depth of fishing operation, and increasing mesh size. There was no clear influence of gear type, average soak time, time of day or effort. Although it is suspected that there are some synergistic effects of mesh size and soak time, with larger mesh sizes soaked for longer periods to target larger fish species which spoil less easily. We estimate over 5,500 marine turtles are captured annually across set nets and demersal longlines in this SSF, more than 4,500 green turtles and 1,000 loggerhead turtles, of which at least 55.5% and 42.8% are expected to be mortalities, respectively. This study provides the first comprehensive assessment of marine turtle bycatch and its drivers within the Northern Cyprus SSF, utilising a multidisciplinary approach to address key knowledge gaps in the Mediterranean for set nets and demersal longlines. Management strategies focusing on set nets, particularly trammel nets, which adequately consider the influence of individual fisher ability, heterogeneity of the fleet, and its operating behaviours will be key to addressing bycatch levels in this fishery which exceeded all previous estimates and those of other Mediterranean set net fisheries. Time of day and depth of fishing operations are clear candidates to explore for mitigation but require careful consideration to balance any trade-offs with other ecologically important species impacted by the fishery, such as elasmobranchs. Satellite telemetry studies focusing on evaluating diel activity patterns, spatial and depth utilisation for life stages present in resident marine turtle populations are required to complement the findings of the current study and ensure proposed targets for mitigation are sufficiently supported.

Incidental captures in fishing gears are one of the greatest threats to sea turtles at sea. In the shallow waters of the Adriatic Sea bottom trawlers are among the most used gear. The Adriatic Sea is one of the two most important neritic foraging areas for Mediterranean loggerhead turtles (Caretta caretta), together with the Tunisian shelf. A high number of turtles are incidentally captured by trawlers every year in the Adriatic waters, a part of which ends up dying because of gas embolism and/or drowning. Reducing the time a turtle is retained in the trawl is essential to reduce both direct and delayed mortality.

To prevent retention in the nets, Turtle Excluder Devices (TEDs) have been designed and enforced in many fisheries of the world. In the Mediterranean, bottom trawlers target multiple species of relatively large size, hence it was thought that TEDs implementation would result in a higher commercial loss compared to traditional gears. Recent tests on a flexible TED provided promising results when used on multi-species trawls. The LIFE MEDTURTLE project (co-founded by the LIFE instrument of the EU) aims to evaluate and show to fishers the efficiency of this type of TED in the Adriatic Sea. Fishers play a key role in sea turtle conservation, therefore, considering their reticence to change the gear traditionally used, it is crucial to determine the optimal gear configuration that can benefit both commercial income and turtle conservation.

TED trials have been conducted by collecting data onboard between November and March of 2021-22 and 2022-23in two known bycatch hotspots in the Adriatic Sea, the area facing Rimini in the North and the Gulf of Manfredonia in the South. In each haul, the number of bycaught turtles, the weight of target (commercial catch) and discard (species that have no commercial value and undersized or damaged individuals) were assessed.

In total, 171 hauls were carried out (66 TEDs, 105 Control, i.e. the regular fishing net) and 15 turtles were captured alive. Standardised Target and Discard (kg per hour of trawling) were analysed with Generalized Linear Mixed Models to examine the influence of Treatment (TED or Control), DN (Day/Night) and Depth, across the various vessels. Target revealed a small but significant decrease with Depth, while Treatment and DN showed no significant effects. For Discard, TED Treatment had a small but significant decrease compared to Control, whereas DN and depth did not have a significant effect.
We conducted a logistic regression to investigate the effects of Treatment, DN, and Depth on turtle bycatch. Treatment significantly decreased the likelihood of turtle bycatch, while DN and DEPTH did not demonstrate significant effects.

These preliminary findings suggest that the use of this flexible TED can be promising in reducing turtle bycatch in the Adriatic Sea. However, they also stress the importance of raising awareness and actively involving the fishing community in conservation efforts for the effective implementation of conservation strategies and the sustainable coexistence of commercial fishing and marine wildlife conservation.

The direct capture of sea turtles is a persistent threat to sea turtles in Southeast Asia. Indonesia remains a global hotspot for the exploitation of sea turtles, despite their protected status in Indonesia. Thanks to the cooperation of various stakeholders and the implementation of conservation programmes, egg harvesting has been reduced at some nesting sites. However, conservation at sea remains a challenge. In particular, the lack of information on sea turtle spatial behaviour and the unknown origin and demography of captured turtles challenge the implementation of at sea conservation.

We conducted a questionnaire survey in two communities in Sumatra in March-April 2022 and May-June 2023, and collected information on sea turtle capture. We also identified the species and measured the carapace of sea turtles (n = 45) captured in the surveyed communities. In addition, we tagged four female green turtles (Chelonia mydas, Cm) with satellite transmitters in June 2023 after they had nested in two rookeries (n = 2 on Bangkaru Island/Aceh, n = 2 on TWP Pieh/West Sumatra) located near the surveyed communities. Our study provides insights into sea turtle capture, the demographics of captured sea turtles and preliminary results on the spatial behaviour of tagged green turtles. The main aim of our study was to inform the management of sea turtle conservation in Sumatra.

Our results indicate that the most commonly captured turtles are green (91%) and hawksbill (65%, Eretmochelys imbricata, Ei), which are the most abundant species in the region. Turtles are captured using nets, spears or directly on the beach. Both incidental and accidental captures were recorded. Carapace measurements indicated that the majority of captured turtles were juveniles and subadults - Cm mean CCL = 75cm (n = 35), Ei mean CCL 53cm (n = 10). Given that juveniles and subadults are being targeted, we are concerned about the future recruitment of sea turtles and the threat of declining numbers of nesting females in the area.

Satellite telemetry data provides evidence of connectivity between green turtle foraging sites. Although the rookeries were about 500 km apart, two turtles from Bangkaru Island and one from TWP Pieh moved to the same area in South Nias. The other tagged turtle from TWP Pieh moved along the Sumatran coast to the Bali area, travelling almost 2,000 km. These patterns are consistent with previous studies on the spatial behaviour of green turtles, which have confirmed movements towards fixed foraging areas where different populations mix. We also used our questionnaire survey to identify areas at sea where turtles are at high risk of exploitation. Combining these data with satellite telemetry, we found that all turtles moved within the high risk exploitation areas. The satellite telemetry results therefore highlight the need for increased protection at sea, as protection of nesting sites is limited to female nesting turtles.