Important information about sea turtles can be gathered from stranded animals on the beaches. Several factors can influence in the stranding of sea turtles, such as plastic ingestion, becoming entangled and drowning in fishing nets, swallowing fish hooks, becoming diseased, or becoming stunned in the cold water during wintertime (Pinedo et al. 1996; Bugoni et al. 2001). By studying these strandings, researchers are better able to monitor potential trends in at-sea threats, and can also gauge changes in abundance through time. In addition, long-term studies can also indicate whether the enhancement and innovation of fishing techniques strongly influence the populations of sea turtles. When integrated over wide spatio-temporal extents, stranding data can provide information about diseases (Balazs 1991; Herbest 1994) geographic ranges, seasonal distribution, and life history, such as life stages and growth rates (Snover et al. 2007; Witt et al. 2007; Tomás et al. 2008; Bellido et al. 2010).

Sea turtle strandings on beaches in the southern Brazil are common, mostly of the species Caretta caretta (Green turtles), Chelonia mydas (Loggerhead turtles), and Dermochelys coriacea (Leatherback turtles) (Pinedo et al. 1996; Bugoni et al. 2001), and the 2 main causes of mortality have been the incidental capture in fishing supplies, trawls, gillnets, and longlines (Pinedo et al. 1996; Areco 1997; Kotas et al. 2004) and the intake of anthropogenic materials, such as plastics and strings (Bugoni et al. 2001). Thus, using the best available data to evaluate those impacts is a challenge of conservation biology to get plausible hypotheses regarding the decline of species of concern. To achieve this, the mitigation of stranded sea turtles needs to be an integral element of management plans for sustainable fisheries. Furthermore, these data can also allow inferences to be made on patterns of by-catch when integrated with careful postmortem examinations such as those done by Bugoni et al. (2001), who found high evidences of anthropogenic debris ingestion (60.5%), including plastic and oil pollution on Green turtle stomachs in southern Brazil.

The South Atlantic Ocean has a high biological production, with economic importance despite the artisanal fishers and industrial uses present in the region (Pinedo et al. 1996; Ministério do Meio Ambiente 2008). This coastal area is highly used by sea turtles as a feeding and development site (Pinedo et al. 1996; Centro de Pesquisa e Gestão dos Recursos Pesqueiros Lagunares e Estuarinos/Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis [CEPERG/IBAMA] 2002; Monteiro 2004). Because fisheries and turtles overlap, incidental catch in fishing nets and gillnet fisheries has been cited as the main cause of sea turtle mortality (Areco 1997; Marcovaldi et al. 2002). There have been a few previous studies with stranded turtles on the beaches of southern Brazil (Pinedo et al. 1996; Bugoni et al. 2001; Estima et al. 2003, 2005), which evaluate different ecological aspects. The present study will fill information gaps for stranding patterns in southern Brazil, register possible external impacts, determine the life stages of stranded sea turtles, and evaluate the temporal trends in numbers for each species found along the beaches.

METHODS

The study area includes a 120-km stretch of beach between Balneário Pinhal (lat 30°14′55″S, long 50°13′47″W) and Mostardas (lat 31°10′52″S, long 50°50′03″W), in the coastline of Rio Grande do Sul (RS) (Fig. 1). We conducted monthly surveys from October 2007 to December 2010 and covered a total of 4320 km of beach by car, at an average speed of 30 km/hr. All sea turtle carcasses that were sighted in the swash zone of the beach were recorded, identified, and measured. The curved carapace length (CCL) of each individual was measured in centimeters with a measuring tape from the notch at the anterior end of the carapace to the posterior end of the last posterior marginal scute (Bolten 1999), which aides in determining the stage of an individual (Dodd 1988; Limpus et al. 1994; Miller 1997; Barata et al. 1998). External impacts (cuts, holes, ropes, and nets) were also registered. To avoid recounts of individuals on previous surveys we used acrylic paint to mark each individual's carapace. We assessed mortality throughout the seasons and cite them according to the southern hemisphere austral seasons: spring (October to December), summer (January to March), autumn (April to June), and winter (July to September).

View Figure

• Download as Powerpoint
• Download Figure

To evaluate the temporal trend from 1992 to 2010 of sea turtle deaths along the coast of RS, we used studies conducted between the years 1992 to 2004 (Pinedo et al. 1996; Bugoni et al. 2001; Estima et al. 2003; Monteiro 2004) and the present study. Pinedo et al. (1996) and Monteiro (2004) conducted their studies along the entire coast of RS. On the other hand, Estima et al. (2005) collected the data on the southern coast of RS, between the Mostardas and Chuí beaches, and the study by Bugoni et al. (2001) and the present study were conducted along the middle coast, between the Balneário Pinhal and Mostardas beaches (Fig. 1). For each study, we calculated an index, as the number of dead sea turtles of each species with a sufficient sample size (Leatherback, Loggerhead, and Green) in each 10 km of beach covered in each study was not homogeneous due to differences in sampling efforts. Finally, a χ² test was used to determine the difference between the expected and observed number of each sea turtles species throughout the 4 seasons. The same test was applied to evaluate if there was a difference between adult and juvenile individuals of each species in the study area.

RESULTS

A total of 640 turtles were reported dead between the beaches of Balneário Pinhal and Mostardas, of which 284 individuals were Loggerhead sea turtles (44.4%), 278 were Green sea turtles (43.4%), 27 were Leatherback sea turtles (4.2%), and 51 were unidentified individuals (8.0%). In addition to these, we documented 10 juvenile individuals near the swash zone of the beach, including 8 Green turtles, 1 Loggerhead turtle, and 1 Hawksbill turtle (Eretmochelys imbricata) (Table 1). A greater number of strandings (Green: χ²  =  85.5, df  =  3, p < 0.0001; Loggerhead: χ²  =  322.5, df  =  3, p < 0.0001; and Leatherback: χ²  =  41.0, df  =  3, p < 0.0001) were observed in the summer and spring than expected for these seasons. Hawksbill turtles did not show significant differences between observed and expected value in the seasons (Fig. 2).

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Sea turtle species life stages and size registered on the coast of Rio Grande do Sul, Brazil, during the period of 2007 to 2010.

View Fullscreen

The greatest amounts of dead Loggerhead, Green, and Leatherback turtles were found in spring, followed by summer, with 84.2% of the total amount of turtles found during these seasons (Fig. 2). Loggerhead turtles were found more frequently in the spring of 2009 (n  =  89), 2007 (n  =  56), and 2010 (n  =  48), whereas Green turtles were more frequently encountered in the summer of 2008 (n  =  68) and spring of 2010 (n  =  66) and 2007 (n  =  48). Most of the Leatherback turtles were found in spring of 2009, 2010 (n  =  7 each year), and 2007 (n  =  6), with most of the stranded individuals found in November (n  =  16). The least amounts of dead sea turtles were found in spring 2008 (n  =  7) and winter 2009 (n  =  8). Between the 2007 and 2010 years, we observed a greater number of juvenile strandings of Green (χ²  =  201.6, df  =  1, p <0.001), Loggerhead (χ²  =  85.4, df  =  1, p < 0.0001), and Leatherback (χ²  =  19.2, df  =  3, p < 0.0001) turtles than expected with respect to each life stage. Hawksbills did not show a significant difference between the adults and juveniles expected and observed value.

Of the 640 sea turtles found on the beaches between Balneário Pinhal and Mostardas, the CCL was measured for 457. The most frequently encountered individuals were juveniles, followed by adults and subadults. Green sea turtles were the most abundant juveniles (n  =  219). For this species, only 1 adult and 6 subadults were documented. Loggerhead turtles showed the highest number of adults (n  =  37), in addition to 170 juveniles of this species. Of the Leatherback turtles, 95.7% were adults and 1 was a juvenile (Table 1). Only 1 Hawksbill sea turtle was found alive in May 2010. This individual was a juvenile that was found in a swash zone of the beach and showed no signs of apparent injuries.

Of the 640 individuals found on the beach between 2007 and 2010, 13 showed impacts from interactions with human activities. One Loggerhead turtle was found entwined in a net, which indicates the direct action of fishing; another one had deep cuts in its carapace; one individual had no carapace; 2 had a 1-m rope around their necks; and 2 were headless. One Green sea turtle was found caught in a fishing net, 2 had partially-removed carapaces, 2 turtles had deep punctures in their carapaces, and 2 were headless. Leatherback turtles did not present any impacts such as cuts, punctures, nets, or ropes.

An analysis of the tendency of sea turtle strandings from 1992 to 2010 shows that mortality of Leatherback, Loggerhead, and Green sea turtles, which are the most abundant species off the coast of southern Brazil, has increased (Fig. 3; Table 2). The highest numbers of beached Loggerhead and Green sea turtles occurred from 2007 to 2010, and corresponded to 0.657 and 0.644 turtles per 10 km, respectively. The number of Loggerhead turtles increased throughout the study, and there was a significant increase in mortality from 1995 to 2004 and from 2004 to 2005. There was also a continual increase in the number of stranded Green turtles from 1997 to 1998. This number of strandings was greater than that found from 1992 to 1995 and from 1995 to 2004.

View Figure

• Download as Powerpoint
• Download Figure

Table 2. Mortality of sea turtles on the coast of Rio Grande do Sul, Brazil, during the years 1992 to 2010. Values in parentheses refer to an index of stranding by species (number of individuals/10 km of beach).

View Fullscreen

DISCUSSION

From 2007 to 2010, there was a predominance of Loggerhead and Green sea turtles strandings (87.8% of all strandings), which is consistent with studies carried out by Pinedo et al. (1996) and Bugoni et al. (2001) along the RS coast. Loggerhead, Green, and Leatherback sea turtles occur regularly in this region, whereas the Olive Ridley (Lepidochelys olivacea) and Hawksbill turtles are only occasionally documented in southern Brazil (Lema 1994; Pinedo et al. 1996; Valls et al. 2011). There has been a low number of stranded Olive Ridley and Hawksbill turtles because of the low abundance of these species in this region. Studies conducted on long-line fishing boats in southern and southeastern Brazil and Uruguay have documented high frequencies of incidental deaths of Loggerhead and Leatherback turtles (Kotas et al. 2004; Pinedo and Polacheck 2004; Domingo et al. 2006; Giffoni et al. 2008; Monteiro 2008). According to Kotas et al. (2004), the fishing area for the fishing fleets is located in the convergence zone of the Brazilian Current and the cold waters from the Falklands Current on the coast of RS, and this area of high productivity coincides with sea turtles' feeding and development locations (CEPERG/IBAMA 2002). Incidental catches in fishing nets, surface and deep-water long-line hooks, and gillnets are the principal causes of sea turtle deaths, followed by shrimp trawling, which is responsible for the highest number of captures and deaths of juveniles, subadults, and adults (Kotas et al. 2004; Pinedo and Polacheck 2004; Tudela et al. 2005). However, the number of dead turtles on the beaches should be considered as a minimal estimate of deaths in the sea, since not all turtles that die at sea reach the swash zone of the beach (Epperly et al. 1996).

The higher rate of stranded Loggerhead, Green, and Leatherback turtles found in spring and summer on the RS coast is characteristic of these species in the region, which was confirmed in this and other previous studies (Pinedo et al. 1996; Lontra 2003; Barata et al. 2004; Monteiro 2004). This fact is due to the migratory nature of these species in this feeding area, and it may be associated with the fishing dynamic of the region (Monteiro 2004) and the presence of the species in warm waters. Sea turtles usually migrate to warmer latitudes in fall and winter to feed off the southern Brazil coast in summer, when the water is warmer from the increased influence of the Brazil Current. Of the 457 turtles found stranded on the beaches between Balneário Pinhal and Mostardas, there was a prevalence of dead juvenile Loggerhead and Green sea turtles (85.6%), which demonstrates a greater occurrence of this stage group in this feeding and development region. On the other hand, juveniles are more susceptible to interactions with human activities, such as ingestion of plastic materials (Balazs 1985; Plotkin and Amos 1990), and many end up dying due to the incidental catch by fishermen (Marcovaldi 1991; Areco 1997; Soto and Beheregaray 1997; Kotas et al. 2004). The average sizes of the Loggerhead sea turtles found on the RS coast were similar in the study by Bugoni et al. (2003) (mean  =  73 cm, minimum [min]  =  63 cm, maximum [max]  =  97 cm, n  =  16), the present study (mean  =  65.4 ± 19.9 cm SD, min  =  15 cm, max  =  120 cm, n  =  207), and in the study by Kotas et al. (2004) of those incidentally caught by pelagic long-lines (mean  =  58 ± 7.7 cm SD, min  = 46 cm, max  =  73 cm, n  =  54). In Argentina, López-Mendilaharsu et al. (2006) also documented juvenile individuals (mean  =  69 cm, min  =  51 cm, max  = 83 cm, n  =  124) in the Rio de La Plata estuary, and Albareda et al. (2003) studied turtles in the coastal region (mean  =  71 cm, min  = 57 cm, max  =  88 cm, n  =  12). The highest number of beached turtles were Green sea turtles, and the most frequently found were juveniles (mean  = 43 ± 14.1 cm SD, min  =  13 cm, max  =  93 cm, n  =  226) (96.9%). Monteiro (2004) and Bugoni et al. (2001) documented similar sizes for this species along the RS coast (mean  =  40.5 ± 5.9 cm SD, min  =  29 cm, max  =  68 cm, n  =  332 and mean  =  37.7 ± 3.1 cm SD, min  =  28 cm, max  =  50 cm, n  =  332, respectively), as did López-Mendilaharsu et al. (2006) and Albareda et al. (2003) along the coasts of Uruguay (mean  =  41.4 cm, min  =  28.0 cm, max  =  79.5 cm, n  =  429) and Argentina (mean  =  39.9 cm, min  =  32.2 cm, max  =  51.0 cm), respectively. These data indicated that the Argentina, Uruguay, and the southern coast of Brazil are a wide feeding and development area for sea turtles, mainly Loggerhead and Green sea turtles with similar stage structures.

On the other hand, the most frequently (95.7%) found adult turtles were Leatherback turtles (mean  =  141.5 ± 27.6 cm SD, min  =  105 cm, max  =  220 cm, n  =  22), with only 1 juvenile found (CCL  =  57 cm). Monteiro (2004) found similar results with individuals stranded along the RS coast (mean  =  136.7 ± 16.4 cm SD, min  =  95 cm, max  =  180 cm, n  =  78), with a greater occurrence of adult individuals, which was also documented in Uruguay (López-Mendilaharsu et al. 2006) and Argentina (Albareda et al. 2003). Subadults and adults have a high reproductive value in a population and are frequently observed in feeding zones in temperate and cold waters, whereas juveniles generally remain in waters with temperatures higher than 26°C (Eckert 2002). However, in tropical and subtropical regions some prevalent oceanographic factors, such as resurgence, systems of currents, and oceanic fronts, favor a concentration of prey distributed throughout shallower waters (i.e., jellyfish), which may alter the diving patterns of the Leatherback turtles (López-Mendilaharsu and Rocha 2009) and contribute to the increase in incidental death caused by fishing interactions and ingestion of plastic.

The Loggerhead and Green sea turtles showed signs of interactions with human activities in the form of pieces of fishing net and cuts or punctures in the carapace, partially removed shell, or absence of the head, which were provoked by sharp objects. Bugoni et al. (2001) documented cuts in the carapace and neck of a Green turtle and 2 Loggerhead turtles and the entire removal of the shell of another individual. Incidental catch in fishing nets has been widely documented (Barata et al. 1998; Kotas et al. 2004; Pinedo and Polacheck 2004). However, it is difficult to determine the cause of death of these individuals because deaths by drowning when the animals are caught in the nets do not generally leave marks on the carcasses (Bugoni et al. 2001), and many dead individuals do not reach the beaches (Epperly et al. 1996). In Uruguay, Lezama et al. (2004) documented 49.3% mortality of sea turtles from deep-water gillnets, which operate in coastal waters of the continental shelf and are responsible for 61% of fish caught (Haimovici 1998). The type of fishing gear used and the location and depth are factors that influence the capture of turtles, and the time remaining in the net in the water is the fundamental factor in the survival of turtles and the avoidance of death by drowning, which depends on the target fish species (Pupo et al. 2006). Although external signs of interaction with human activities were not observed in the carcasses of Leatherback turtles found on the beach, recent studies indicate incidental catch by fisheries as the principal cause of death for this species (Thomé et al. 2003; Barata et al. 2004; Lewison et al. 2004), in addition to the ingestion of plastic and other debris (Lutcavage et al. 1997; Bugoni et al. 2001).

From 1992 to 2010, there has been an increasing tendency in the mortality of Loggerhead, Green, and Leatherback turtles along the RS coast. The number of turtle strandings increases as the number of impacts from human activities increases in the feeding and development zones. These impacts include plastic materials in the water, oil spills, and increased fishing efforts in response to higher demand for protein and reduced fishing stocks (Haimovici 1998). Fishing is one of the causes of the incidental mortality of sea turtles and has likely been underestimated in this study due to the low amount of external impacts from interactions with human activities documented. However, interactions with human activities tend to increase as the Brazilian government encourages the fishing industry through subsidies and financial incentives to increase the number of fishing trips and quantity of catch (Brasil 2009a, 2009b) for many species that are already overexploited. These 3 species of sea turtles are globally threatened: Loggerhead and Green sea turtles are endangered and the Leatherback turtle is critically endangered (International Union for Conservation of Nature 2012). To reverse the population decline, international and local forces are needed to implant management plans in areas used by turtles to feed, develop, and reproduce, in a way that will avoid habitat degradation and incidental catch by fishing activities.