Study Area

Guinea-Bissau is a small West African country with a generally flat coast indented by numerous flooded valleys and estuaries (Fig. 1). A large continental shelf extends well beyond the coast and the numerous islands. Most of the mainland coast is characterized by wide intertidal areas with mudflats and sandbanks and extensive mangrove areas. Only in the extreme north and south can one find sandy beaches facing the open ocean. The Bijagós archipelago is made of 88 islands and islets, of which only 21 are permanently or temporarily inhabited. Sandy beaches with deep water approaches are uncommon, being mostly found on the outer face of the outer islands. Most of the Bijagós are sparsely populated, with a total population of ca. 25,000 people in a land area of 90,000 ha (Cuq 2001). Besides resident populations, which live mostly on subsistence agriculture, there are increasing numbers of seasonal immigrants (mostly from Senegal) who fish from large open pirogues, mostly targeting sharks, rays, barracudas, snappers, and other large predatory fish. There is no industrial fishing within the archipelago or near the mainland shore. The climate is tropical, hot, and humid. Rainfall is highly seasonal, with a monsoonal type rainy season from May to November that, in coastal areas, averages around 2000 mm per year.

View Figure

• Download as Powerpoint
• Download Figure

Field Surveys

Most of our detailed knowledge on the sea turtles of Guinea-Bissau results from a number of field surveys that are described below in some detail. Surveys for green and hawksbill turtles generally occurred during peak nesting activity in August–October, whereas those for olive ridley turtles occurred during the peak nesting period in January–February.

Orango National Park Survey

Between December 1992 and November 1994, frequent surveys of the 6 most important beaches of the Orango group of islands, more recently classified as the Orango National Park, were carried out (Fig. 2). Beaches were walked in the early morning, up to 5 times per month, with an overall full coverage of all the months of the year. Overall, 320 beach surveys were carried out. The mean number of surveys per individual beach is 53.3 ± 15.5 (range 26–71). Tracks and nests were identified and noted, making a distinction between “fresh” (< 24 hours) and “old”. Tracks were erased to avoid posterior counting. The fate of old nests was also noted as intact or depredated (in which case, the type of predator was noted). To obtain a broad estimate of the total number of nests laid per beach and per month, the mean number of “fresh” nests recorded per visit was multiplied by the number of days of that month. Values from different beaches and months were then added to obtain a general estimate for the Orango group.

View Figure

• Download as Powerpoint
• Download Figure

Poilão Nesting Surveys

Surveys of nesting beaches to assess sea turtle numbers coming ashore are usually based on track counts. On Poilão, 2 extensive surveys were carried out in 2000 (from 19 July and 14 December; Catry et al. 2002) and in 2007 (from 27 June to 18 November). Tracks were counted daily in the early morning. For various reasons, there were 1-day gaps of coverage in 23 (15% of the period covered) and 11 (8%) scattered dates, respectively in 2000 and 2007. Values for days with no counts were estimated by linear interpolation (Catry et al. 2002).

Because not all adult female emergences result in clutch deposition, it is important to estimate the proportion that results in egg laying. Around Poilão, and at different times throughout the study period in 2000 and 2007, we randomly chose 75 turtles emerging from the sea and followed them from a distance (without disturbance) until they laid their clutch (on 61 occasions) or returned to the ocean without laying. Laying success was thus estimated as 61/75  =  81.3% (95% confidence limits: 70.6%–89.4%). The estimations of numbers of nests were obtained by multiplying the laying success by the numbers of tracks counted. Less extensive surveys were carried out from 2001 to 2006 using a similar methodology, including from 3 to 93 daily counts per year in the months of July–October.

To summarize data on interannual variation of laying effort on Poilão we calculated a nest population index by the following procedure. The year 2000, for which we have the most complete and reliable counts, served as a baseline for all other comparisons. We calculated the mean daily number of tracks by periods of 10 or 11 days (21–31 July, 1–10 August, 21–31 October 2000), and arbitrarily set the index value of the year 2000 as 100. We then applied the formula

Where Xi₂₀₀₀ is the mean number of tracks on the relevant i 10–11 day period in the year 2000 and Xi_a is the corresponding value for the focal year (a). We opted to use this index rather than estimate total nesting numbers per year given that some year estimates were based on a reduced number of counts, leading to a large uncertainty in total numbers present.

For the study of the nesting biology on Poilão, we used standard methodologies, such as marking nests and following their fate until hatching (see Catry et al. 2002 for more details). In 2007, the high density of nesting resulted in a massive loss of nest markers, which prevented any rigorous estimate of nesting success. To obtain an idea of the damage caused by digging females on other turtle nests, we checked for evidence of damage around 20 laying females per night, during each night in August, September, and October 2007.

Green Turtle, Chelonia mydas

Status and Distribution

The green turtle is the most common sea turtle in Guinea-Bissau. Its reproduction has been confirmed in virtually all the Bijagós islands. Furthermore, it nests in small numbers along the mainland coast (Fig. 1). The initial 1990 survey (Limoges and Robillard 1991b, 1991c) and subsequent work indicate that, by far, the most important nesting site is located at Poilão (Fig. 3).

View Figure

• Download as Powerpoint
• Download Figure

Results from the 1992–1994 survey of the 6 main beaches of the Orango Group are presented in Table 1. Low intensity monitoring in other years further confirmed those results, suggesting that a few hundred nests are laid annually in the Orango National Park. A survey of those same 6 beaches in early January 2008 revealed the presence of at least 374 shells of individual green turtles that, according to local witnesses, had been killed and consumed during 2007. Of these, 91% were found at a beach that had an estimated 22% of the total number of nests laid during the 1992–1994 survey. From these data, one can reasonably speculate that during the peak year of 2007 (see data on interannual variability on Poilão) there were a few thousand nests laid in the Orango National Park.

Table 1 Results from the December 1992–November 1994 beach surveys in the Orango National Park. “Old” nests were more than 24 hours when found and “fresh” were from the previous night. Estimates of numbers of nests for the whole park were made by extrapolating from the numbers of fresh nests seen to days without surveys (see methods for more details).

View Fullscreen

In the year 2000, 9733 tracks were counted at Poilão, and it was estimated that 7397 clutches were laid in this season (Catry et al. 2002). Subsequent surveys with variable field effort show that this was not an exceptional year, but rather a normal one (Fig. 4).

View Figure

• Download as Powerpoint
• Download Figure

The total number of tracks from 27 June to 18 November 2007 was estimated at 33,991. Multiplying this number of tracks by 1.05, to account for the period of the year when no counts were carried out, and by 0.813, to correct for laying success, we obtain an estimate of 29,016 nests (95% CI  =  25,197–31,907) laid on Poilão in 2007.

Foraging Areas

Foraging sea turtles are frequently seen around many islands of the Bijagós Archipelago and even off some of the mainland beaches, for example, near Varela. Some areas must have a very high density of young turtles, which is suggested by the capture of 19 green turtles in only 3 short capture sessions near Unhocomo and Unhocomozinho, in May and September 2003 (Table 2).

Table 2 Measurements of sea turtle samples from Guinea-Bissau. Samples from Poilão and Adonga are from nesting females while those from the Unhocomo group are from juveniles (a possible adult male with a CCL of 91 cm was excluded from this sample) captured for research purposes. Measurements of samples from the North Coast were from shells obtained in coastal villages from Varela to Jeta (data from Dontaine et al. 2001). CCL  =  curved carapace length; CCW  =  curved carapace width.

View Fullscreen

Olive Ridley Turtle, Lepidochelys olivacea

The olive ridley turtle appears to be the second most numerous species in Guinea-Bissau, although its distribution may be less wide than that of other species, as we only have confirmed nesting records from the Orango National Park. Results from the 1992–1994 survey of the Orango group are presented in Table 1. By far, the most important nesting beach was on the barrier island of Adonga, with almost half of the nests recorded. Limited surveys in January–March 2008 suggest these broad levels of nesting are still maintained, despite significant human predation, particularly at Adonga.

In Guinea-Bissau, olive ridleys are solitary nesters, with a peak nesting season in the central part of the dry period (Fig. 5). Mean clutch size (± SD) for 7 nests at Adonga was 130.3 ± 14.2 eggs (range  =  109–150 eggs). During the 1992–1994 survey in the Orango Group, 35% of the old nests (n  =  142) when first detected had suffered predation. In 26% of the cases predation was by humans, 52% by crabs, and 22% by monitor lizards.

View Figure

• Download as Powerpoint
• Download Figure

Hawksbill Turtle Eretmochelys imbricata

Reports of hawksbill turtles and their shells can be found sparsely all over the Guinea-Bissau coastal zone, indicating a wide at-sea distribution. In the 1990 survey, 5 fresh tracks were recorded on Poilão and 2 on Adonga. This hinted at the presence of a sizeable population in the Bijagós, but posterior data did not confirm so. Few nests were found in the Orango group in 1992–1994 (Table 1). During the intensive Poilão survey in 2000, 4 turtles were seen ashore and 6 nests identified (Catry et al. 2002) while in 2007, 7 different individuals were recorded, one of them twice. It must be stressed that hawksbills females and their tracks are likely to be often overlooked amongst the large numbers of green turtles at this site. In any case, the total numbers nesting annually at Poilão must be in the low tens. An educated guess (based on data presented here and other personal observations) would be that there are 50–200 nests being laid per year in Guinea-Bissau. We are unsure about the source of the estimate presented in Spotila (2004) of 200 nesting females per year in this country, but under our present state of knowledge, this value would seem too high, given that each female should lay several clutches.

According to reports from locals and observations on Poilão and Orango, hawksbills have a peak nesting season in July–October, but there are also several dry-season records, for example, in January and March. Mean clutch size (± SD) for hawksbill nests at Poilão was 163.8 ± 29.7 eggs (range: 142–215 eggs, n  =  5).

Loggerhead Turtle Caretta caretta

Loggerheads are very rare in Guinea-Bissau. At least one old carapace originating from the Bijagós was positively identified in the 1980s (Limoges 1989; Boesl 1995), but it is not known if this individual was nesting locally. In 2003, during detailed enquires on the island of Unhocomozinho, we were told by local collaborators that turtles of a rare species had nested on 6 occasions, on 2 different beaches, between 28 February and 9 May 2003 (from the dates, at least 3 different individuals must have been involved). These collaborators knew other turtle species well, gave a description that was consistent with loggerheads, and furthermore positively identified the species when shown published images.

Leatherback Turtle, Dermochelys coriacea

Leatherbacks are rare on the beaches of Guinea-Bissau. They have been recorded nesting on the mainland, at Varela, and on the islands of Unhocomo, Orango, Adonga, Orangozinho, Canhabaque, and João Vieira. Surveys of the Orango group (Table 1) highlighted Adonga as the main nesting site. An educated guess for the present national nesting levels would point to 25–75 nests per year. Information from several sources all confirm that in Guinea-Bissau the species lays during the dry season, from November to April, with a single known record from June.

Status and Biology

The present review highlights the importance of Guinea-Bissau as a nesting area for sea turtles. Data presented here give the first insight into the annual variation of nesting numbers on Poilão and confirm previous information showing this site to be of outstanding importance in an Atlantic context. The number of green turtle nests laid annually on Poilão varies between ca. 7000 and 29,000, which possibly makes this site the third most important for this species in the whole Atlantic, after Tortuguero, Costa Rica (Bjorndal et al. 1999) and Ascension Island (Godley et al. 2001), and places it amongst the top 12 sites in the world (see Spotila 2004). We should note here that we regard our estimates of nest numbers on Poilão as conservative (particularly the 2007 estimate). The very high density of turtle tracks makes their counting difficult and on several occasions we counted more turtles ashore during the night than tracks in the following morning. Hence, it is not unlikely that nest numbers were above 30,000 in the latest study year.

Broderick et al. (2006) quote the Poilão population as an example of a positive green turtle demographic trend, but this seems to be based on only 2 (relatively imprecise) data points. Considering the annual variability in nesting numbers typical of this species (Broderick et al. 2001), such assessment seems to be premature.

Green turtle hatching and emergence success at Poilão was high in a year (2000) with “typical” nesting density (Catry et al. 2002). In 2007, with a much higher density, it is clear that many nests were destroyed by laying females. The partial destruction of a nest can result in rotting eggs that negatively affect both the original and the overlapping clutch (Caut et al. 2006). Knowing that approximately one third of the nesting turtles had scattered eggs around them in 2007, it seems likely that nesting success was significantly lowered by density-dependent factors (see Bustard and Tognetti 1969; Caut et al. 2006), which also suggests that the potential for population growth at Poilão is limited by the availability of suitable nesting habitat.

The presence of small numbers of leatherbacks nesting in Guinea-Bissau is not surprising, given that the species has been recorded reproducing in almost all West African countries (Turtle Expert Working Group 2007). Nesting populations of hawksbill and olive ridley turtles may also be of international importance, at least in a West African context. Nesting loggerhead turtles are perhaps more unexpected. If confirmed to nest regularly, this would represent an extension of the known nesting range of the species in the Eastern Atlantic.

Body sizes, internesting intervals, and clutch sizes measured for green, hawksbill, and olive ridley turtles are within average values for those species (see Hirth 1997; Miller 1997; Weir et al. 2007). Nesting seasons also seem to be similar to other locations in the northern hemisphere for green and hawksbill turtles (Richardson et al. 1999; Spotila 2004; Pérez-Castañeda et al. 2007), which in Guinea-Bissau nest mostly during the rainy season. However, olive ridley turtles display a pronounced nesting peak in the middle of the dry season, which contrasts to what is found, for example, in Angola or in the eastern Pacific (Plotkin 2007; Weir et al. 2007), but is similar to the situation found in India (Spotila 2004). A similar pattern may apply to the leatherback, which in Guinea-Bissau seems to avoid nesting during the rainy season, while elsewhere in the Atlantic it often nests in periods with considerable rainfall (e.g., Caut et al. 2006; Weir et al. 2007). Such inter- and intraspecific variability in nesting seasons is interesting in the context of temperature-dependent sex determination. Given that pivotal temperatures for sex determination seem to be a conservative trait in sea turtles (Wibbels 2003), one might not expect such strong variability of the timing of nesting in a geographical narrow range such as the Bijagós archipelago. More research is needed to understand the differences between the reproductive seasonality of different species and implications for hatchling sex ratios.

Along the Atlantic coast of Africa, fibropapillomatosis has been confirmed to occur in green turtles from Gabon and equatorial Guinea (Formia et al. 2007), and turtles showing large tumors likely to be fibropapilloma have been recorded in Guinea-Bissau (3 cases recorded by us from the year 2000) and neighboring countries, such as Gambia and Senegal (Barnett et al. 2004). This disease has a very high prevalence in some areas and there are concerns that it may have a significant impact on sea turtle populations, particularly green turtles (Aguirre and Lutz 2004). Our results from Poilão suggest that the incidence of fibropapillomatosis is currently extremely low in the adult green turtle population, which may be linked to the fact that the turtles live in waters relatively free of sources of pollution and is an indicator of good ecosystem health in their foraging and mating grounds (Aguirre and Lutz 2004).

Threats and Conservation Initiatives

Interviews carried out all over the coastal area of Guinea-Bissau in the last 2 decades clearly indicate that there is a general feeling that sea turtles have decreased in numbers within living memory. It is easy to obtain verbal reports stating that turtles used to be numerous on specific nesting beaches where they are now scarce or almost absent.

The Bijagós people maintain many of their traditional beliefs and their life is governed by numerous taboos and rules, even if this may be rapidly changing. A rather strict taboo involves the remote islet of Poilão, which can only be visited during occasional and rare ceremonies of social and religious significance. This could justify its favorable conservation status. Elsewhere, there do not seem to be any general traditional rules protecting sea turtles, although at some places we have been told that leatherbacks can only be killed under special circumstances (see also Bernatets 2005).

In the Bijagós, marine turtles can be used in religious ceremonies (e.g., Bernatets 2005), but their meat is not reserved for special occasions and can be consumed at any time. In fact, the consumption of turtle meat and eggs seems to occur very frequently and, on most nesting beaches, few if any turtles or nests are left alone when found by locals. The data from the Orango National Park, where at least 374 adult female green turtles were killed for consumption in 2007, somewhat illustrates the magnitude of the problem. On the other hand, there is very little use of turtle shell for manufacture of goods.

Marine turtles are strictly protected by the national fisheries law, although there is no specific mention to their eggs and nests in the law text. Turtle meat or eggs are very seldom seen on markets anywhere in the country and most harvesting results in local consumption. Nevertheless, there are recent reports of turtles being exported alive by foreign fishermen. Targeted capture at sea is only known to be done by nationals on a small scale, around Unhocomo and Unhocomozinho. Whether there are occasions and places where foreign fishermen specifically target turtles in Guinea-Bissau waters is unknown, but we have first-hand reports of this happening not far from the national border, in Senegalese waters, which, given the mobility of fisherman across frontiers, makes it likely that it happens in Guinea-Bissau too.

Off Guinea-Bissau, many industrial fishing trawlers operate, coming from countries as diverse as China, Spain, France, Italy, and Portugal. There are no impositions from the national government to use turtle excluder devices and many fishing vessels operate illegally. Very little is known of the impact of the industrial fishing fleet in Guinea-Bissau waters. An early study suggested that between 500 and 1000 turtles could be caught per year (Limoges and Robillard 1991c), but there is no information on the methods used or on the reliability of this estimate. A latter (superficial) assessment based on 11 interviews with national fisheries observers and on information on the number of licenses for the international fleet, indicated that the annual catch might be on the order of 300 turtles, of which many could be released alive (Broderick and Catry 1998).

Sea turtles are regularly captured in monofilament nets set from open boats, in coastal waters, to catch predatory fish, such as sharks, rays, barracuda, jacks, and snappers. In Guinea-Bissau, such fishing is mostly carried out by foreign fishermen (generally from Senegal, Guinea-Conakry, and Sierra Leone), but also by nationals, often in association with foreigners. Turtles may drown in the nets or, if caught alive, they are generally killed and consumed. We have no quantitative data on the level of this type of mortality, but it is almost certainly very high, as informal contacts with fisherman indicate that turtles are frequently captured and that one large fishing canoe can, at certain times and locations, capture several turtles per day. According to the villagers of settlements near important nesting beaches in the Orango National Park, in the past 2–3 decades, the capture by foreign fishermen of very numerous turtles was witnessed and, as a result, nesting turtles (particularly olive ridleys and perhaps hawksbills), which were previously abundant, became scarce in the area. Green turtles, being captured in nets less often (partly because they nest at a season with comparatively little fishing activity), were less affected and their populations did not decline so much, despite important harvest on the nesting beaches.

Sea turtle nesting beaches and adjacent mating grounds were taken into consideration when planning the creation of a network of national protected areas (e.g., Limoges and Robillard 1991a; Catry et al. 2002). Conservation efforts resulted in the recognition by the UNESCO of the Bolama-Bijagós Biosphere Reserve that covers the entire Bijagós archipelago, within which 3 protected areas have been formally created by the national government. The most important sea turtle nesting sites in Guinea-Bissau are situated in the core zones of those protected areas. The main direct effect of the creation of the protected areas was the removal, since 2004, of large semipermanent foreigner fishing camps and the prevention of further temporary or permanent settling. Furthermore, many environmental education campaigns and conservation messages have reached the Bijagós villages through the community radio.

On the other hand, few enforcement activities have taken place. There have been virtually no penalties for people found to have killed sea turtles or harvested eggs. This results mostly from a situation in which investments in protected areas have been mostly directed toward community-based development initiatives and where enforcement activities have often been seen as contrary to the principles of protected areas comanagement. This, and the insufficient human and financial resources for patrolling and vigilance, mean that Guinea-Bissau parks share the fate of most of their West African counterparts, where most megafauna are seriously threatened for these very same reasons (e.g., Oates 2002). However, the attitudes are changing, even if slowly. In the last 2–3 years, there have been some fishing parties fined for illegal fishing in the core areas of marine parks, which may help deter the use of such areas. Nevertheless, the illegal use of nets by foreigners and nationals in those critical sectors is still common practice. In the core areas of the Orango National Park, for example, illegal fishing boats were present every day when surveys were carried out in January–March 2008.

The regular presence of monitoring and research teams in Poilão has certainly reinforced the traditional protection that benefits this site. Without such presence, it is likely that temporary camps of foreign fisherman would be established and the illegal harvest of turtles would take place as happened in the past.

The development of ecotourism activities in the Bijagós may contribute to raise the interest of both local populations and government officials in the conservation of the area and its natural resources, demonstrating that the conservation of marine turtles can promote, rather than prevent, the sustainable development of the area. For this to effectively happen, it is essential that local communities derive direct benefits from such initiatives.