Insights afforded by this work
This study confirms that two species of marine turtles regularly come to the Atlantic coastline of Ebodje to nest. In quantitative terms, nesting within the Gulf of Guinea varies for each of these species. For example, nesting can either be in high nesting concentration zones, but rather a dispersion of regular nesting sites, in small quantities, with an insular predominance. For over 4 years, we observed the number of nests at approximately 237 on a number of sites along the Atlantic coastline of Ebodje, which was less similar to that of Punta Oscura/Punta Santiago in the south of Bioko (with approximately 191 nests for just 2 seasons; 1996–1998) and far less than that in São Tomé, which had a total of 277 nests during the 1998/1999 season (Fretey et al, 2001). The most visited beaches were in the southern zone compared to the northern zone. Historical nesting in the area equally showed predorminance of olive ridley to leatherback turtles; 1999, 30 nests of olive ridley and 12 nests of leatherbacks and 2000, 22 nests of olive ridley and 9 nests of leatherbacks (Angoni 2005; Angoni et al. 2010).
Comparison between current and historical data
The Lepidochelys olivacea population declined in the South Atlantic due to frequent accidental capture of adults of both sexes by shrimp trawler boats and is a priority for conservation in West Africa (Fretey 2001). Each female adult and nest are therefore of great importance for the survival of this species. Thus, there are concerns that leatherbacks have two hot spots on each side of the ocean: the Surinam-French-Guiana region and the region of south Gabon-North Congo. Each of these spots has multitude sites of average and minor importance, with female turtles not necessarily remaining faithful to one site, i.e., in Central Africa, a female leatherback normally nesting on Mayumba beach (South Gabon) could occasionally nest on Conkouati (North Congo), or in Equatorial Guinea, at São Tomé to the south of Bioko or near Ebodjé. Due to the decline of the species in the Pacific Ocean, the D. coriacea population "linked" to the hotspot in the Eastern Atlantic has become the most important in the world, together with that of the Western Atlantic, and is therefore an international priority for its preservation. The small random sites, as this seems to be the case for the beaches around Ebodjé, must therefore have the best protection possible, which is not the case at the moment because the females are still being killed at sea and on land.
There is a very clear peak of turtles going up on the beach between November, December and January. There are fewer turtles going up on beaches in the northern zone. The busiest beaches are those in the southern zone; Ipenyendje, Bekolobé, Likodo, Ebodje, Mbendji and Bouandjo, to a lesser extent beaches such as Elombo and Lolabé. It must be noted that the backs of the southern beaches are the most occupied by human habitations in the zones covered by this study. The average CCL/CCW and clutch size reported by this study were closer to those reported by previous researchers (Table 3 and Table 4). It is most likely young turtles that start to nest during this season since the sexual maturity size is close to 130 cm on the pictures that were taken, and there is no amputation of the postcentral spur that might explain why they are small.
Table 3
Comparison of curve-line length and width of dorsal shells for female olive ridley turtle nesting in various regions
Places | Average (cm) | Extremes (cm) | N | Source |
Costa Rica (Nancite) | 63.7 | 57.2–71.4 | 404 | Cornelius and Robinson (1984) |
Costa Rica (Naranjo) | 65.2 | 57.0-72.5 | 53 | Cornelius (1976) |
Mexico (Oaxaca) | 62.98 | 54.0–67.0 | 81 | Frazier (1983) |
Surinam (Eilanti) | - | 64–72 | 14 | Pritchard (1969) |
Surinam | 68.5 | 63–75 | 500 | Reichart (1993) |
Cameroon | 70.86 | 64.5–82 | 29 | Fretey (1999) |
São Tomé | 70.13 | 62–80 | 417 | Fretey et al. (2001) |
These results also confirm that L. olivacea predominates on D. coriacea. It is biologically normal since leatherback turtles, with their skin not protected by keratinous skin and their pseudo shell lacking bony protection, can easily hurt themselves; all females heading onshore to nest avoid rocky areas. Because of its size, heaviness and amplitude of the movements required to move onshore, this species also chooses very wide beaches devoid of obstacles. On a beach close to the
Table 4
Comparison of curve-line length and width of dorsal shells for female leatherback turtles nesting in various regions
Places | Average length (CCL) | Number | Average width (CCW) | Number | Sources |
Brazil (Espírito Santo) | 159.8 ± 10.5 (139–182) | 24 | | | Thomé et al. (2007) |
French Guiana (Ya:lima:po) | 154.6 ± 6.7 (127–252 SCL) | 1,328 | 87.3 ± 6.21 | 1,328 | Girondot and Fretey (1996) |
Surinam | 154.1 ± 6.7 (128–184) | 1,840 | 113.2 ± 5.0 | 801 | Hiltermann and Goverse (2007) |
Trinidad (Matura Beach) | 157.6 (139.7–210.0) | 104 | - | - | Chu Cheong (1990) |
Costa Rica (Tortuguero) | 156.2 ± 10.6 (124.0-180.3) | 35 | - | - | Leslie et al. (1996) |
Bioko | 156.06 ± 14.87 | 458 | - | - | Formia et al. (2000) |
Gabon (Gamba) | 150.4 ± 7.6 (130–172) | 819 | 108.3 ± 6.6 (126–144) | 819 | Verhage et al. (2006) |
village, when going onshore and emerging from the water, this species will be afraid of any moving outline. However, when it is in an oviposition, several humans will be able to move around it with flashlights. Nothing will interrupt the nesting protocol. In contrast, ridley turtles move a lot and are very fast. They are also protected by a scaly skin and hard shell and move easily between rocks to reach the beach. Their nesting sites do not need to be very wide, but as they are fearful, ridley turtles instead dig their nest on the verge of or in vegetation areas where they can hide. These behavioral characteristics explain why the occupation of the two species is different and why leatherback turtles mostly nested in the Likodo-Ebodjé-Mbendji area during this season.
Regarding tracks for this present study, there were 2 predominant groups, 170 ± 32 cm and 68 ± 8 cm, that were attributed to D. coriacea and L. olivacea, respectively. Their analysed data were limited to D. coriacea: 85–207 cm and L. olivacea: 44–100 cm. The majority of the tracks were mostly female L. olivacea. A set of alternate tracks belonging to C. mydas was spotted in Ebodje and Ipeyengue, heading under the trees. The species has never been observed nesting in Cameroon but nesting 200 km south of Bioko (Tomas et al. 2010). Ridley turtles typically nest twice during a season with a renesting interval of 12 to 36 days. Each nesting included an average of 110 eggs (usually between 80 and 150 eggs). In the data collected during this study, nesting 160 eggs is a credible standard. The clutch including 178 eggs is more enigmatic. For leatherback turtles, the annual cycle is rare. The cycle is usually 2, 3, 4 or 5 years. Leatherback turtles nest an average of 87 ± 18.9 eggs for each clutch. According to age, size, and health condition for each female leatherback turtle, the number of clutches by season can vary between 7 and 12 (Fretey, 2005).
Body size effects
The geographic distributions of the body size in each of the species have different or similar sizes in the area or colonies elsewhere. We suggest several environmental and evolutionary explanations for this result, which make different predictions and oceanic productivity. For instance, the Carribean and West Atlantic colonies tend to contain large turtles, whereas those in the Mediterranean Sea, North Indian Ocean and eastern Pacific are typically small. Some of the reasons responsible for this variation include size-biased methods of human harvesting, heavy/extensive exploitation of adults and changes in nutritional differences (Boulon and Frezer 1990). Other explanations may include heavy beach predation on eggs and hatchling (poaching), which favour females to produce large numbers of eggs, which correlate with increasing body size, because body size is positively correlated with clutch size and clutch frequency. The trade-off between clutch size and egg size was confirmed for marine turtles after factoring out the effects of body size. Large olive ridley species tend to lay large eggs, which give rise to large hatchlings.
This study has clearly shown that clutch size increases with the body size of individual olive ridley turtles nesting on the coastline of Ebodje. This is in line with many previous studies that have explained the variation in clutch size by curved carapace length/width (Bjorndal and Carr, 1989; Hays and Speakman, 1991; Hays et al., 1993). Marine turtles grow throughout their lives, and a correlation between body size and reproductive output has been demonstrated for several species (Moll 1979; Iverson 1992b; Wallis et al. 1999). Additional resources can be allocated to make larger eggs, to produce more eggs, or to both. In D. coriacea, larger females produced larger eggs, but body size showed a lesser effect on clutch size when compared to the other species (larger females did not lay more clutches). This result is in keeping with findings elsewhere (Frazer and Richardson, 1986; Johnson and Ehrhart, 1996). Several chelonians show a trade-off between egg size and clutch size (Elgar and Heaphy 1989). For instance, their clutch size was not correlated with egg size, but after correcting for the effect of body size on egg size, a weak correlation was shown between egg size and clutch size; clutch size decreased with increasing egg size. Larger female turtles allocate more resources to reproduction, and these resources are mainly invested in larger eggs. Larger eggs produce larger hatchlings, and larger hatchlings have a higher chance of survival than smaller hatchlings (Janzen 1993). When observing nesting in females, we noticed on one occasion that the diameter of the egg was too large to pass through the caudal gap between the carapace and plastron. However, the range in clutch size would probably vary less if clutches were counted multiple times for individual renesting females (Price et al. 2006); such counts could not be performed in this study, as we never registered any number of renesting individuals. Comparing the average clutch size of this study (77 ± 20.1) and 72 ± 10.6 for the corresponding CCL and CCW, respectively, to the study of Reina et al. (2000) finding an average clutch size in Pacific leatherbacks of 64.7 ± 1.4, this study supports the finding that Pacific leatherbacks have a smaller clutch size than Atlantic leatherbacks. In addition, the CCL range of 132–150 cm found in this study compared with 144–147.6 found by Reina et al. (2000) also supports that Pacific leatherbacks tend to be smaller despite the smaller sample size for this study.
Nesting trends
The nesting pattern for marine turtles in the Ebodje nesting zone is similar to that of several rookeries in the region (Tomas et al. 2010) and worldwide (Chaloupka, 2001). They nest all year round with a distinct peak in beaches of the southern and northern zones. However, there have been some differences in the timing and duration of the nesting season in the zones. Such differences have equally been reported on the Island of Bioko (Garcia, 1996) and Sao Tome and Principe (Loveridge and William, 1957), where the Gulf of Guinea’s turtle stock has a breeding peak during the dry season as opposed to the wet season for the other stocks (Bowen et al., 1992; Chaloupka, 2001). In addition to the mature green turtles choosing Bioko as their place for nesting, immature green turtles often swim in the coastal waters of Cameroon, Equatorial Guinea, Sao Tome and Principe and Gabon for foraging. Formia et al. (2006) equally reported the consideration of two sites as individual nesting populations due to geographical separation. In the Indian Ocean, Stepler 2006 also reported breeding differences in the season timing of green turtles in two geographically close locations. Individuals nesting by olive ridleys and leatherbacks on beaches in the Ebodje spawning zone used different feeding grounds in Bioko and elsewhere. There have been reports of tagged individuals nesting on two sites, Tromelin and Europa, that were found feeding off the east and west coasts of Madagascar (Le Gall, 1988).
The negative correlation between species size, nesting effort and the length of the different beach zones could be explained by the existence of stressors/threats on the species. Additionally, marine turtles do not nest in consecutive years; for instance, green turtles nest on average for three to four years (Carr and Carr, 1970; Le Gall, 1988; Hays, 2000; Solow et al., 2002), 2–3 years for olive ridley and 2, 3, 4 or 5 years for leatherbacks. After a breeding season, individuals migrate back to their feeding sites where they spend most of their time feeding and accumulating enough fat reserves to sustain vitellogenesis and undertake the next migration to the breeding grounds (Limpus and Nicholls, 1988; Kwan, 1994). For remigration, intervals largely depend on the environmental conditions at the feeding sites (Hays, 2000; Solow et al., 2002). Some of the reasons for this shift include good or bad feeding conditions (Carr and Carr, 1970; Hays, 2000; Solow et al., 2002), sea surface temperature (SST) positive anomalies and sources of seagrass nutrition (Solow et al., 2002; Limpus and Nicholls, 1988; Hays, 2000; Solow et al., 2002; Chambers et al., 1999).
The general increase in the number of tracks and the annual decrease in the species mean size for this study when compared to historical data confirms that the protection of nesting sites might have negatively impacted the nesting populations (Chaloupka and Limpus, 2001; Balazs and Chaloupka, 2004; Troëng and Rankin, 2005). Previous researchers have reported how important nesting sites for marine turtles have been negatively affected by human settlement, with turtle eggs and meat being harvested (Hoareau, 1993; Frazier, 1975). Additionally, poaching, poverty and bycatch in artisanal fishing were additional reasons to account for this negative trend in nesting populations. The high interannual variability in the number of tracks reported in this study and the different environmental conditions at their nesting/feeding sites were probably due to different seasonality. This causes difficulties in the interpretation of population trends (Le Gall, 1988; Hays, 2000; Solow et al., 2002).
Insights into threats
Records from the Ebodje local poachers and their fishing communities revealed widespread marine turtle exploitation, with mainly their meat, eggs, oil and shells used for local subsistence. Exploitation has increased since the early 1990s due to increased poverty levels and human habitation in coastal areas resulting from dense urbanization and an increase in the coastal population of neighbouring villages. Species vulnerability to threats is due to their life history (slow growth rates, late maturity, and high longevity), physiological attributes and behavior, which make them extremely sensitive to environmental changes. Turtle bycatch was identified as a source of mortality and hence a primary driver of population declines in marine turtle species (Lewison et al. 2004a).
Men from the whole of the region will not hesitate to walk many miles at night to find a turtle that has gone up onto the beach to nest. The most isolated beaches are subject to poaching. A machete or a large knife is enough to kill and cut up this defenseless sea animal, which is slow to move on land. Meanwhile, immature/juvenile green and olive ridley turtles entangled in coarse netting and a leatherback turtle caught in submerged monofilament netting, as in Njifonjou et al. 1995). Catching a leatherback turtle in a net is considered by the net owner catastrophic because the turtle makes enormous holes that are difficult to repair. The priority is therefore given to saving fishing equipment. The turtle, which is too heavy and bulky, cannot be brought back to the village, as the Iyassa pirogue is too small and unstable. The fishermen also fear being taken out to sea by the still living leatherback, and there is no lack of dramatic stories recounting such events (real or imaginary). Back on land, the meat is stripped off the female killed while nesting. At this time, the neighbors come with bowls and wheelbarrows, and the meat is shared between families. The frequency of interactions (defined as accidental encounters with artisanal fishing gear that can result in injury and possibly death) depends on spatiotemporal overlap between critical habitats for a given species. Information on bycatch rates, the amounts of fishing effort on which these rates are based, and the status of the affected population(s) is crucial to characterize bycatch patterns and predict the potential impact of bycatch (Lewison et al. 2004b; Soykan et al. 2008).
Artisanal fishers in Ebodje based their choice of fishing gear on a variety of factors, including the fisher’s knowledge, ease of use, initial costs, yield and seasonality. Traditionally, most fisheries elders trained community members to use the fishing gear that they are approved for. They also passed on knowledge of wind tides and current to improve fishing efficiency and decrease the likelihood of injury or loss at sea. Such a knowledge base heavily influenced the fishermen’s choice of fishing gear and techniques. Accidental bycatch of turtles in artisanal fishing gear was widespread. Green turtles were the most bycaught species, with sporadic olive ridley and leatherback turtle bycatch reported and occurring year-round. Most of the 15 turtle carcasses recorded (juveniles and adults of both sexes) showed evidence of bycatch, such as net marks. Interviewees identified certain foraging/nesting areas that coincided with long-line fishing areas. Fishing gear such as gillnets and longlines caused selective mortality among these older age classes in sea turtles (Lewison et al. 2004b).
Conservation Implications
This study clearly demonstrates the value of broad application of survey efforts in (i) identifying the spatial distributions of marine turtles (protected species) in areas where there is potential for interactions with industry developments and coastal village activities and (ii) estimating seasonal changes in density in ways that may be considered effective management of species. To date, conservation efforts for these populations have been concentrated on nesting sites. Access on beaches of the Ebodje spawning zone is restricted and continuously controlled, and there is no harvesting of females and eggs on the beaches. Females nest in an environment fairly undisturbed by humans, unlike other sites where the presence of humans has hampered nesting activities and caused a decline in population (Bertrand et al., 1986). Efforts have been made to maintain the environmental conditions of the different beaches as optimal as possible for the four visiting turtle species that are vulnerable to habitat destruction (DIREN, 2003; Seminoff, 2004a). For all four marine turtle species, conservation measures are most effective when they ensure long-term nesting success, hatchling survival and adult protection from poachers (Balazs and Chaloupka, 2004; Dutton et al., 2005; Troëng and Rankin, 2005). Nesting success and hatchling survival need to be determined at the Ebodje beaches because of the possible decline experienced by the population. Sea level rise may also influence nesting success as they change beach configuration, which could destroy nests that have been laid. However, the direct impacts of sea level rise on nesting success have not yet been studied at these sites.
Environmental conditions at the feeding sites are equally important to the juvenile green turtles and hawksbill turtles as they determine the nutritional capacity of the individuals to sustain vitellogenesis and migrate to the nesting sites. Thus, it is important to know where a population nests and feeds to establish the best conservation management possible. Off-nesting and breeding site conservation plans need to be implemented for marine turtle populations in the Ebodje spawning zone. The soft sandy beaches and rocky locations of the marine coastal environment of the Atlantic coastline of Ebodje are good nesting and feeding sites for these turtle populations, and illegal hunting (poaching) still occurs. Regulations are not well inforced in the country due to its large size and high poverty (Berthin and Harding, 2005). More sites of nesting/foraging by female turtles need to be identified so that further studies can provide additional information on the population status. The main problem with the conservation of marine turtles is the time lag of conservation measures impacts due to the generation length of the species (Mortimer, 1995; Seminoff, 2004a). The current impacts of humans on the dynamics of the Ebodje nesting and foraging population are unknown. The general increased number of tracks at the different beach zones for this study when compared to historical data suggests that the adult population is recovering and has not been greatly disturbed at the nesting/feeding sites. However, the current situation will determine the future status of the species; thus, effective conservation efforts must be pursued along with strategies adapted to the regional context.
Conservation activities
Our results generally reinforce previous studies, which found that long-term conservation efforts have a strong positive association with species conservation status (Hays 2004, PLTRT 2006). In this study, we showed that the continuous presence of specific conservation activities was associated with lower threat impact level assessments to the beach sections.
Wildlife law enforcement
Reduced levels of law enforcement are known to expose marine turtle populations within our study sites to increased poaching pressure and other illegal activities. Thus, a continuous presence of patrollers will improve law enforcement patrols policies, which are associated with lower impact levels of threats to the different zones.
Beach cleaning
Increases in beach erosion and degradation as a result of sea level rise due to climate change are causing the fall of coastal beach vegetation, making it impossible for turtles to reach nesting areas.
Ecotourism and research
Introducing secondary conservation activities such as turtle-based ecotourism and research can have indirect effects on wildlife preservation (Vieitas et al., 1999). The success of a valuable ecotourism program is based on local participation of the community in educational programs, tour-guide training, festivals, and the employment of former egg poachers to patrol the beaches and protect the nests (Marcovaldi and Marcovaldi, 1999). A significant contribution offered by this is the deterring of poachers, thereby creating ‘‘wildlife refugia’’ (Campbell et al. 2011, Laurance 2013), in raising employment opportunities and public awareness of the value of conservation, therefore favoring a lower threat impact level (Clem and Clevo 2002).
Lack of funds/grants
Shortage or limited funds will signify the stop of beach surveys and the resume of direct harvest of adults (Anonymous 2000), with records of turtles transported by land from Ureca to Luba and Malabo markets of ca. 250 green turtles, 50 leatherbacks, 6 olive ridleys, and 6 hawksbills per season from south Bioko beaches. It is equally possible for funds to support research and conservation projects.