Introduction
By-catch is the fraction of the catch unintentionally captured during a fishing operation in addition to the target species. It may refer to the capture of other marketable species that are landed, to commercial species that cannot be landed (e.g., undersized, damaged specimens), to non-commercial species, or incidental catches of endangered, vulnerable, or rare species (e.g., sea turtles, chondrichthyans, marine mammals) (FAO, 2015; FAO, 2016). Longline fishing gear has a higher by-catch rate than any other type of commercial fishing gear. Longline (pelagic longline [PLL]) fishing mainly affects pelagic sharks and rays, but also benthic species (bottom longline). Pelagic sharks are caught as by-catch by longline fisheries targeting swordfish and tunas (Oliver et al., 2015); in the Mediterranean, the most common by-catch pelagic shark species are the blue shark (Prionace glauca), the basking shark (Cetorhinus maximus), the great white shark (Carcharodon carcharias), the smooth hammerhead shark (Sphyrna zygaena), and the common thresher shark (Alopias vulpinus) (FAO, 2016; Carpentieri et al., 2021). The status of such pelagic sharks has worsened over time (Walls and Dulvy, 2021), partly due to by-catch overfishing (Lewison et al., 2004). Pelagic sharks can play important roles in marine ecosystems through many mechanisms; hence, their decline may initiate trophic cascades and affect the overall community structure (Stillwell and Kohler, 1982; Heithaus et al., 2008; Ferretti et al., 2010; Ward and Myers, 2016). Within the Mediterranean basin, the highest by-catch rate for PLL was found in the Alboran Sea, followed by the Adriatic Sea (34.3% and 15.1% of the total catch, respectively) (Megalofonou et al., 2005). Therefore, in this study, the Adriatic Sea was targeted as an important area for the by-catch of pelagic sharks (Bartolí et al., 2017). The blue shark is the most caught species by PLL, representing over 70% of the elasmobranch catch (Bradai et al., 2012), and it is considered a critically endangered species in the Mediterranean by the International Union for Conservation of Nature (IUCN). Sea turtles are another non-targeted species impacted by longline fishing in the Mediterranean Sea. Indeed, both pelagic (e.g., Deflorio et al., 2005; Báez et al., 2007; Casale et al., 2007; Jribi et al., 2008; Cambiè et al., 2010) and bottom (e.g., Casale et al., 2007; Jribi et al., 2008) longlines show a consistent number of sea turtles caught by this gear (Casale, 2011; Carpentieri et al., 2021). Three sea turtle species are present in the Mediterranean: the leatherback turtle (Dermochelys coriacea; Dermochelyidae), the green turtle (Chelonia mydas; Cheloniidae), and the loggerhead turtle (Caretta; Cheloniidae). However, few leatherback turtles enter the Mediterranean from the Atlantic without breeding in the basin (Casale et al., 2003). The other two species present in the Mediterranean are more strongly impacted by PLL fishing activity as by-catch (Encalada et al., 1996; Casale, 2011).
Measures to mitigate by-catch are needed; however, due to the variety of factors influencing the interactions of fisheries with elasmobranchs and other endangered species, simple solutions for mitigation are not available. Importantly, PLL by-catch encompasses several species bearing different tolerances to the impact of (a) being caught or (b) being susceptible to being caught in the first place. Moreover, by-catch mitigation might also be more complex for sharks because some species (e.g., blue shark, shortfin mako, and common thresher shark) might be landed and marketed with a lower value than the target species (swordfish and tunas) (FAO, 2016).
By-catch mitigation strategies typically include modification of the gear and/or of the fishing strategy. To date, longlines are the gear with which most efforts have been made to reduce shark by-catch, probably because longlines have a large impact on by-catch globally (Gilman et al., 2016). The longline gear changes include the distance between floats (to adjust the depth at which the hooks fish), the material of the leaders or branch lines (e.g., wire or nylon), the type of bait, and the shape and size of hooks (Gilman et al., 2016). Effects on target species catch rates are frequently taken into account in by-catch mitigation research because fishers are more likely to adopt mitigation strategies that do not result in a reduction in landed target catch (Hall et al., 2007; Ward and Hindmarsh, 2007; Campbell and Cornwell, 2008). For instance, Beverly et al. (2009) showed that removing shallow hooks from PLL gear reduced catch rates of epipelagic species like endangered sea turtles, while maintaining the catch rates of targeted tunas. In contrast, shorter longline soak times result in lower levels of unwanted catch mortality; however, they may also result in lower catches of the target species (Ward et al., 2004; Carruthers et al., 2009).
There have been opposing findings regarding the impact of hook shape on by-catch. Traditionally, J-shaped hooks (J-type hooks) have had the reputation of resulting in a high by-catch of protected, endangered, or threatened species (PET species), and replacing them with circle hooks (C-type hooks) has been shown to be especially effective in reducing the by-catch of marine turtles (Piovano and Gilman, 2017). Moreover, the use of C-type hooks is considered a relatively low-cost by-catch mitigation tool; therefore, a large number of studies exist globally about the C-type hook effect on by-catch (e.g. Kim et al., 2006; Pacheco et al., 2011; Afonso et al., 2012; Godin et al., 2012; Gilman et al., 2016). Overall, the effects of C-type hooks on by-catch species seem to be species- and area-specific (Pacheco et al., 2011; Gilman et al., 2016). However, due to the small number of studies in the Mediterranean and opposing findings in other regions, there is a need to conduct further research to find effective by-catch mitigation strategies that consider multiple factors and potential trade-offs among species (Lucchetti and Sala, 2010; Piovano et al., 2010). As a result, the objective of this preliminary study is to compare the effects that use of traditional J-type hooks and C-type hooks used by a swordfish fishery in the South Adriatic Sea on target and by-catch species.
Discussion
Several new laws have been adopted in recent years. The primary goal of European Union (EU) Regulation 2019/1241, which has replaced and integrated EU Regulation 812/2004, is to minimize and, whenever possible, eliminate incidental catches of sensitive species to ensure that fishery-related mortality does not jeopardize those species’ conservation status. A new binding recommendation of the General Fisheries Commission for the Mediterranean (GFCM/44/2021/16) adopted in 2021 on additional conservation and mitigation measures for the conservation of elasmobranchs in the Mediterranean Sea asks Contracting Parties, inter alia, to “require fishing vessels, catching accessorily and incidentally sharks species, to limit by-catch of sharks”, such as blue sharks, and “to improve the conservation status of elasmobranch species, mitigate and where possible eliminate the risk of incidental taking of elasmobranch in fishing operations and the associated mortality “. A new “EU action plan: protecting and restoring marine ecosystems for sustainable and resilient fisheries” (2023) also calls on member states to “swiftly take measures to (…) protect sharks” and a new GFCM “Regional Plan of Action to monitor and mitigate interactions between fisheries and vulnerable species in the Mediterranean and the Black Sea” is in preparation. In the Mediterranean, some by-catch reduction devices (BRDs) (e.g., grid, ultraviolet LED, circle hooks) have been tested to mitigate the impact that several items of fishing gear, such as passive net (Virgili et al., 2018; Lucchetti et al., 2019), trawl net (Lucchetti et al., 2016; Lucchetti et al., 2019), and PLL (Piovano et al., 2009; Piovano and Swimmer, 2017), have on sea turtles. However, the effects of BRDs on sharks have been tested less in the Mediterranean than in other areas (Carpentieri et al., 2021; Bradai et al., 2022).
Measures to mitigate the impact of PLL by-catch include adaptations of the gear, fishing area, bait, soak time, and setting and hauling times of the longline (Bigelow and Maunder, 2007; Coelho et al., 2012; Gilman et al., 2016). The adaptations of the gear include the distance between floats, which affects the hook fishing depth (Afonso et al., 2012), the material of the leaders and/or branch lines (e.g., wire or nylon) (Afonso et al., 2012), the use of repellent (e.g., chemical or magnetic) (Lucas and Berggren, 2022), and hook size and shape (Piovano and Gilman, 2017). The gear modification that is the most commonly tested is hook shape (e.g., Kim et al., 2006; Piovano et al., 2009; Ward et al., 2009; Sales et al., 2010; Pacheco et al., 2011; Afonso et al., 2012; Coelho et al., 2012; Godin et al., 2012; Gilman et al., 2016; Piovano and Gilman, 2017; Guo et al., 2022) – probably for being a relatively low-cost by-catch mitigation measure (Gilman et al., 2016).
Circle hook
Although the substitution of J-type hooks with C-type hooks has been tested in different areas, the effects of this substitution on both target species and by-catch species are not univocal (Gilman et al., 2016). In some cases, the C-type hook was found to have no effect on the catchability and mortality of either the target or by-catch species (Ward et al., 2009; Afonso et al., 2012; Coelho et al., 2012; Godin et al., 2012). In other cases, the C-type hook influenced the quantity of the by-catch species caught (Kim et al., 2006; Piovano et al., 2009; Sales et al., 2010; Gilman et al., 2016) or influenced the mortality of the by-catch and/or target species (Pacheco et al., 2011; Guo et al., 2022). These results from the scientific literature suggest that the effects of replacing the J-type hook with the C-type hook may be related to many factors, such as the species, area of fishing and/or season, bait type, water temperature, hook size, and/or depth (Gilman et al., 2016; Guo et al., 2022). Thus, several aspects should be taken into consideration to better assess the effects of C-type hook use on target and by-catch species, such as possible trade-offs and potential conflicts between different species; the use of a C-type hook can have beneficial or adverse effects, depending on the species (Sales et al., 2010; Pacheco et al., 2011; Piovano and Swimmer, 2017). The sample size of fishing sets in the present preliminary study is small; nevertheless, the results contribute to filling an information gap, especially in the Mediterranean basin, where few studies of this type exist (e.g., Piovano et al., 2009; Casale, 2011), by showing some of the positive effects of C-type hooks on the capture condition of blue sharks (discussed in detail below). Notably, further research is needed to strengthen the present results and to better understand the influence of factors such as area of fishing and/or season, bait type, water temperature, and hook size on the by-catch in the Mediterranean Sea.
Effects of a C-type hook on swordfish
The CPUE and size of swordfish caught by the two types of hooks did not differ significantly in our experiment. These findings are in accordance with the results obtained in another Mediterranean area (the Strait of Sicily) (Piovano et al., 2009) and the equatorial Atlantic Ocean (Coelho et al., 2012). In contrast, a significant decrease in the capture rate of swordfish was detected when using C-type hooks along the Brazilian coast (Sales et al., 2010). These apparently contradictory results, in addition to reaffirming the influence of the area on the catch-effectiveness of the two hook types, indicate that several factors play a role (e.g., bait, season, position of bait on the hook), and it is not easy to disentangle them to solely evaluate the effect of the hook (Broadhurst and Hazin, 2001; Watson et al., 2005; Pacheco et al., 2011; Coelho et al., 2012). Thus, repeating our experiment in the same area, with the same boat, crew, and bait – and in a limited time frame (seven fishing sets) – could aid in separating the other synergistic effects of the hook effect on the catchability of both target and by-catch species. However, the limited period of time could mask some potential effects from other factors (e.g., water temperature and fishing season). In terms of catch condition, swordfish did not show any significant difference between hook types, and these results are in accordance with Coelho et al. (2012) in the equatorial Atlantic Ocean. In particular, in the present study, the soak time was between 10 and 20 hours, which can induce death due to physiological stress (Carruthers et al., 2011). Therefore, the hook shape’s positive effect on mortality could be masked by the soak-time effect (Carruthers et al., 2011). Altogether, our results suggest that the use of the C-type hook instead of the J-type hook did not affect the total number or size range of the targeted specimens (i.e., the swordfish) that were caught.
Effects of a C-type hook on blue sharks
In this study, blue shark CPUE and the size of specimens did not show any significant difference between the hook types tested. This result is in accordance with Ward et al. (2009) in Australia, Pacheco et al. (2011) in the Atlantic Ocean, and Coelho et al. (2012) in the equatorial Atlantic Ocean. Along the Brazilian coast, Sales et al. (2010) found that the use of C-type hooks increased the catch rate of blue sharks, whereas Kim et al. (2006) found that C-type hooks decreased the catch rate of blue sharks in the eastern Pacific Ocean. Again, these contrasting results could be the consequence of synergistic effects due not only to the type of hook but also to several other factors (e.g., area, fishing practices, season, bait type, and soak time). Interestingly, in a meta-analysis compiling 15 studies, Godin et al. (2012) found no significant quantitative effects of the C-type hook on blue shark catchability. For the evaluation of the impact of hook type on blue shark catchability, the methodology applied in our study – which tends to decrease variability beyond the type of hook as much as possible – can be very useful (Carruthers et al., 2011; Godin et al., 2012). In addition, we observed that hook type significantly impacted the haulback condition of blue sharks. Indeed, the blue sharks caught by C-type hooks were in significantly better condition than those caught by J-type hooks. The C-type hook is supposed to increase this likelihood because both target and by-catch species are typically hooked in the jaw with this hook type (Ward et al., 2009); in contrast, the J-type hook is reported to become lodged in deeper locations (e.g., throat, esophagus, or stomach), which damages the internal organs and thus increases catch mortality (Falterman and Graves, 2002; Cooke and Suski, 2004; Watson et al., 2005; Carruthers et al., 2009; Afonso et al., 2011; Pacheco et al., 2011). Post-release survival is positively and directly correlated with the haulback condition of captured pelagic sharks, such as blue sharks (Campana et al., 2016; Musyl and Gilman, 2018). Therefore, C-type hooks – although they do not affect catchability – positively affect the haulback condition, which in turn affects post-release survival (Musyl et al., 2009; Musyl and Gilman, 2019; Whitney et al., 2021). Thus, using C-type hooks and releasing by-catch species could be an effective management measure to reduce fishing mortality and improve the conservation of shark populations (Bowlby et al., 2021; Knotek et al., 2022). This needs to be confirmed by a larger sample size in the Mediterranean Sea and the inclusion of other factors, such as seasonality, water temperature, bait type, and depth. Notably, although CPUE is an important indicator of catch rates, there is also the possibility that some species (targeted or not) bite the hook and escape (i.e., “bite-offs”) and are, therefore, not included in the calculation of catch rate by CPUE, especially when dropline is composed of monofilament (Afonso et al., 2012). The use of C-type hooks could also benefit the survival of these animals, especially blue sharks, because they are designed to hook fish in the mouth rather than in the gut, reducing the chances of gut hooking with internal injuries and thereby increasing the chances of survival after hooking (Afonso et al., 2011). Hence, blue sharks that were hooked with C-type hooks and managed to free themselves have a better chance of survival compared to animals hooked internally with J-type hooks. This assertion is supported by the present study’s findings that blue sharks caught with C-type hooks were generally in better condition than those caught with J-type hooks. This should be better encompassed in future studies comparing the effects of different hook types.
Effects of a C-type hook on loggerhead turtles
In our experiment, the CPUE, size, and haulback conditions of loggerhead turtles did not differ significantly in relation to a hook shape. Unlike our results, a significant reduction in the catch rate and deep-hooking of loggerhead turtles caught by C-type hooks was found in the Atlantic Ocean (Sales et al., 2010) and the Mediterranean Sea (Strait of Sicily) (Piovano et al., 2009). These differences in our experiment could be due to the low number of loggerhead turtle specimens caught by each hook type (Kim et al., 2006). Moreover, the catch conditions of the loggerhead turtles were good in most cases because the branch lines were long enough to allow the hooked specimens to reach the surface and breathe.
Effects of a C-type hook on pelagic stingrays
Similar results to those for the loggerhead turtles were found for the pelagic stingrays; there were no significant differences in CPUE, size, or haulback condition between the two hook types. In contrast, previous studies (e.g., Carruthers et al., 2009; Ward et al., 2009; Pacheco et al., 2011) found significant differences in the catch rate and catch condition. The lack of significant differences in the present study could be due to the low number of specimens caught.