Sea turtles are found throughout the world's oceans and play an integral role in ecological processes that maintain the health of the oceans. They provide keystone services by maintaining strong and diverse seagrass beds and coral reefs (Meylan 1988; Leon and Bjorndal 2002; Bjorndal and Jackson 2003; Eckert and Hemphill 2005). Sea turtles also help sustain a balanced trophic structure that benefits economically important fish species (Houghton et al. 2006; Lynam et al. 2006). Moreover, they act as a bridge in energy flow between pelagic and coastal ecosystems by transporting nutrients onto beaches in the form of eggs and hatchlings (Bouchard and Bjorndal 2000; Lovich et al. 2018). Energy from sea turtle eggs helps maintain coastal dune habitats by providing nutrients to dune vegetation and supporting nest predator populations (Bouchard and Bjorndal 2000; Lovich et al. 2018). All 7 species of sea turtles are globally threatened with extinction due to a range of anthropogenic sources of mortality (Gibbons et al. 2000; Lovich et al. 2018). Successful conservation and protection of these species will require a full understanding of their global distribution and nesting status; however, there remains a significant gap in our knowledge of the status of sea turtles in many parts of the world (Richardson 1999; Weir et al. 2007; Wallace et al. 2010). Here, for the first time, we report on sea turtle nesting patterns in Ghana, West Africa.

Researchers rely on standardized nest surveys to assess the population status of sea turtles since obtaining demographic information from aquatic life stages is challenging (Richardson 1999). For example, long-term nesting data were used to document the rapid decline of eastern Pacific leatherback turtles (Dermochelys coriacea) (Spotila et al. 2000) and the recovery of Hawaiian green sea turtles (Chelonia mydas) (Chaloupka et al. 2008; Kittinger et al. 2013). Nesting data are also used to identify anthropogenic disturbances, such as the reduction in nesting activity due to artificial lighting (Mann 1978; Witherington and Martin 2000) or declines in hatching success and recruitment due to invasive nest predators, such as coyotes, domestic dogs, and pigs (Ruiz-Izaguirre et al. 2015; Engeman et al. 2016).

Consistent nesting beach surveys also provide data necessary to understand recovery potential by detailing important life history features, such as individual reproductive output, lifetime reproductive potential, age at reproduction, and survivorship (Bjorndal et al. 1999; Broderick et al. 2003; Balazs and Chaloupka 2004). These parameters are important for recognizing stochastic variation in nesting activity from trends that illustrate a decline or an increase in nesting population size (Ceriani et al. 2019). Furthermore, such data can demonstrate how sea turtles respond to natural weather events (i.e., cyclones) (Limpus and Nicholls 1988; Chaloupka 2001) and changing environmental conditions associated with climate change: ocean temperatures (Solow et al. 2002), sea level rise (Fish et al. 2005), and warming beach temperatures (Fuentes et al. 2010). Resource managers rely on these data to develop effective recovery strategies that function to reduce sea turtle mortality and increase recruitment from nesting beaches.

The eastern Atlantic Ocean and western Africa coastline are now recognized as an important region for sea turtle conservation. The IUCN/SSC Marine Turtle Specialist Group considers the area a high-threat, high-risk area containing 10 of the 58 recognized Regional Management Units (Wallace et al. 2010). Five of the world's 7 species of sea turtles utilize the region for foraging, migration, breeding, and nesting (Formia et al. 2003). Juvenile turtles use foraging grounds throughout the region (Ferreira et al. 2018). Cabo Verde supports the third-largest loggerhead rookery in the world with over 20,000 nests laid annually (Marco et al. 2013). Bijagos in Guinea-Bissau is an important green turtle nesting site with over 40,000 nests being deposited in a single season (Catry et al. 2009; Agyekumhene et al. 2017). The largest leatherback nesting rookery in the world is in Gabon, with around 15,730–41,373 estimated breeding females (Witt et al. 2009). Nesting occurs throughout the region, but unfortunately, there is a paucity of long-term nesting data from many portions of the region.

The status and threats to sea turtles in Ghana are poorly documented due to financial and logistical constraints that have prevented the development of a consistent long-term assessment program (Allman and Armah 2008). Early records report that hawksbill (Eretmochelys imbricata), leatherback, and green sea turtles utilize the beaches for nesting (Irvine 1947); however, Loveridge and Williams (1957) also include the olive ridley (Lepidochelys olivacea). More recent reports indicate that loggerhead (Caretta caretta), hawksbill, green, leatherback, and olive ridley sea turtles nest along Ghana's coastline (listed in increasing order of abundance) (Hughes 1988), but there have been no longitudinal studies conducted to assess nesting activity across seasons. We therefore initiated this study to assess nesting patterns and conservation requirements for sea turtles in Ghana. This study presents the results of systematic nesting surveys conducted on a nesting beach in Ghana, West Africa, over an 8-yr period. We also provide a description of current threats and management activities. Reporting nesting data across species and years helps inform management strategies that function effectively toward species recovery (Foin et al. 1998; Wallace et al. 2010).

METHODS

Study Site. — Mankoadze is a small fishing community located immediately west of the Muni-Pomadze Ramsar Site in Central Region, Ghana (Fig. 1). The Muni Lagoon supports a mangrove estuary of brackish water that serves as a nursery for economically valued fish and is also an important overwintering spot for migratory shorebirds (Attuquayefio 1999; Ntiamoa-Baidu and Gordon 1991; Ntiamoa-Baidu and Owusu 1998). The primary economic activity in Mankoadze is artisanal fishing that takes place year-round. The shoreline is composed primarily of sand with a short section of rocky outcrop. The beach exhibits evidence of erosion with small escarpments forming each year. Vegetation along the back beach is sparse and is composed primarily of coconut palms. The beach is littered with trash, predominantly plastic.

View Figure

• Download as Powerpoint
• Download Figure

Experimental Design. — The survey team conducted sea turtle nesting surveys from November 2012 to March 2020 on a 3.7-km stretch of beach that extends from Mankoadze toward the Muni Lagoon. Surveys were not conducted on national holidays and in dangerous weather conditions. We conducted surveys each year from 20:00 PM to 4:00 AM between 1 November and 15 February. Beginning in 2016, we also carried out daily morning surveys from 16 February to 31 October because some nesting activity had been reported during those months.

We recorded each emergence as either a successful nest (deposition of eggs) or a nonnesting emergence (NNE) (no eggs were deposited). The location of each activity was recorded using a Garmin eTrex GPS unit (Garmin Ltd, Olathe, KS). We used a flexible measuring tape to measure the curved carapace length (CCL, notch to tip) and curved carapace width (CCW, maximum width). We installed monel metal flipper tags (National Band & Tag Co, Newport, KY) on the trailing edge of each front flipper for all hard-shelled turtles and inconel metal flipper tags (National Band and Tag Co.) directly through the skin between the tail and each rear flipper for all observed leatherbacks. PIT tags (BioMark, Boise, ID) were injected in the right shoulder of all species. The survey team measured the turtles during oviposition and installed tags after the turtle had completed depositing eggs to ensure that these activities did not interrupt the nesting process. If the turtle was observed before depositing eggs, 1 team member would count the eggs while they were being laid. Successful nests were left unmarked so as not to attract poachers or disturb the fishing activities on the beach.

Morning surveys were conducted to document tracks by species and distinguish nests from NNEs. Nests were identified by the presence of a mound of thrown sand that covered the primary body pit, the presence of a secondary body pit, and/or evidence that sand was thrown using all 4 flippers. A track was considered an NNE if there was no sand disturbance and the turtle turned around without body pitting or if the wall of the primary body pit was still present.

We monitored a subsample of nests (olive ridley, n = 70; leatherback, n = 37; green turtle, n = 5) from 2017 to 2020 to document the incubation period and fate of the eggs. The survey team excavated each nest 3 d after the primary hatching event or after 70 d from deposition to determine hatching and emergence success as described by Miller (1999). Hatching success was calculated as the portion of total eggs that were found empty (greater than 50% intact), and emergence success was calculated as the portion of the total number of eggs that were found empty minus the number of live and dead hatchlings that remained in the egg chamber; both are expressed as a percentage of the total eggs deposited by the turtle. Descriptive statistics were used to characterize the nesting activity and reproductive descriptors for each species across the survey period. We used Excel (Microsoft, Redmond, WA, 2016) and JMP (SAS Institute Inc, Cary, NC, 1989–2019).

RESULTS

A total of 693 olive ridley, 149 leatherback, and 32 green turtle emergences were documented over the 8-yr survey period. These activities resulted in 630 olive ridley, 137 leatherback, and 17 green turtle nests. We observed 35% (2015) to 56% (2012) of all activities and installed flipper tags on 249 olive ridleys, 61 leatherbacks, and 14 green turtles. Across all species, the tag return rate was 2.5%.

Olive ridleys were the most commonly documented species with a mean number of 60 (range = 31–109, ± 26 SD) nests documented annually. The mean number of leatherback nests recorded annually was 17 (range = 3–40, ± 15 SD), while the mean number of green turtle nests observed was 2 (range = 0–5, ± 1.7 SD). We did not encounter any green turtle emergences during the first 2 yrs of the study. Regression analyses indicate a decreasing number of nests each year for olive ridleys (R² = 0.53, p = 0.04), but no patterns emerge for leatherback nesting (R² = 0.11, p = 0.4) or green turtle nesting (R² = 0.43, p = 0.08) (Fig. 2).

View Figure

• Download as Powerpoint
• Download Figure

Olive ridley nesting occurred from September through February, with 96% (605 nests) being recorded during those months. Olive ridley nesting activity was observed each month of the year with the exception of April and May (Fig. 3). Sporadic leatherback nesting activity was observed in September and October, but regular nesting activity occurred from the middle of November through the beginning of February. The months of December and January account for 72% (100 nests) of all documented leatherback nests. Green turtle nesting activity was limited to October through December, with only an average of 1–2 emergences per month (Fig. 3).

View Figure

• Download as Powerpoint
• Download Figure

The mean curved carapace length was 67 cm (range = 56–78 cm, ± 2.9 SD) for nesting olive ridleys, 143 cm (range = 131–158 cm, ± 6.2 SD) for leatherbacks, and 95 cm (range = 88–102 cm, ± 10.5 SD) for green turtles (Table 1). No significant differences in size were found among nesting seasons (p > 0.05 for all species). The mean clutch size was 106 eggs (range = 41–145 eggs) for olive ridley nests, 77 eggs (range = 43–124 eggs) for leatherbacks, and 128 eggs (range = 81–155 eggs) for green turtles. With a mean of 72 d (range = 57–86 d), the leatherback nests had the longest incubation period. The mean incubation period for olive ridley nests was 65 d (range = 43–73 d) and was 63 d (range = 56–69 d) for green turtles. Leatherback nests had the lowest hatching success with an average of 65% (range = 14%–93%), whereas the average for olive ridley nests was 73% (range = 8%–98%) and 74% (range = 67%–83%) for green turtles (Table 1).

Table 1. Summary data for Lepidochelys olivacea, Dermochelys coriacea, and Chelonia mydas nesting in Mankoadze, Ghana.

View Fullscreen

DISCUSSION

This is the first longitudinal study of sea turtle nesting activity in Ghana and demonstrates the regular occurrence of olive ridley, leatherback, and green turtles during a primary nesting season from September through February. Olive ridleys appear to nest sporadically throughout the year. Nesting by loggerhead has recently been documented in Ghana (Allman et al. 2015), but none were recorded during this study in Mankoadze.

Olive ridleys are the most commonly observed species in Ghana; however, nesting in Mankoadze has declined by 46% since the highest nesting season of 2013. This nesting population is part of the East Atlantic Regional Management Unit (RMU) with larger nesting populations ranging from Guinea-Bissau to Angola (Fretey 2001; Wallace et al. 2010). Although nesting data in other regions indicate that populations are decreasing or stable, long-term data are currently not available to assess trends in the eastern Atlantic (National Marine Fisheries Service and US Fish and Wildlife Service [NMFS and USFWS] 2020). Populations in this region are vulnerable to a growing commercial and artisanal fishing industry as well as direct harvest of nesting females (Girard and Breheret 2013; Allman et al. 2020). Fishers in Ghana generally agree that olive ridleys are not as common as they once were (Agyekumhene et al. 2014; Alexander et al. 2017). The lower nesting density observed in Ghana compared with other parts of West Africa may be a function of the amount of increased coastal development and intensity of fishing in the country (Weir et al. 2007; Allman et al. 2020) as well as the high rate of poaching and egg predation by feral dogs (Agyekumhene et al. 2014).

The small number of leatherback and green turtles that nest in Mankoadze may represent depleted nesting populations from historical activity when population sizes were higher. Fishers across different communities report that leatherbacks and green turtles are not seen nesting as often as they once were (Alexander et al. 2017). The leatherbacks of West Africa fall within the Southeast Atlantic and Northwest Atlantic RMUs (Wallace et al. 2010) and the Southeast Atlantic DPS (NMFS and USFWS 2020). Leatherback nesting occurs from Mauritania to Angola and includes the world's largest known nesting population located in Gabon (Fretey et al. 2007; Witt et al. 2009). Leatherback populations in the Atlantic are declining and considered at either a moderate or a high extinction risk (NMFS and USFWS 2020). The green turtles of West Africa are poorly resolved into the Eastern Atlantic and South Central Atlantic RMUs due to limited evidence of population structuring and poorly understood regional migratory patterns (Castroviejo et al. 1994; Wallace et al. 2010). Nesting occurs from Mauritania to Angola with primary nesting sites in Guinea-Bissau, Bioko Island, Equatorial Guinea, and Sao Tome and Principe (Seminoff et al. 2015). Sea turtle populations in the region appear to be declining due to fishery interaction, direct harvesting of nesting females, degradation of nesting habitats, and marine debris and pollution (Catry et al. 2009; Agyekumhene et al. 2014; Aruna 2014; Allman et al. 2020). However, some locations near Ghana, such as Cote d'Ivoire and Benin, have observed increasing nesting activity (Dah et al. 2020; Dossou-Bodjrenou et al. 2020).

With few exceptions (see Loureiro et al. 2011), the mean clutch sizes in Ghana are smaller than those reported from other locations in West Africa (Weir 2007; Catry et al. 2009). The mean clutch size for olive ridleys in other West African nations range from 110 to 130 eggs per clutch and from 80 to 100 for leatherbacks (Kouerey Oliwina et al. 2020). Clutch size is often correlated with body size in oviparous reptiles, and this has been observed for sea turtles (Broderick et al. 2003). Da Silva et al. (2007) reported a positive relationship between clutch size and carapace length for olive ridley turtles nesting in Brazil. A similar pattern has been documented for leatherbacks (Rostal 2015). The nesting turtles in Ghana appear to be at the smaller end of their range in size compared with individuals from other beaches in the region but similar in size to those nesting in the Republic of Congo and the Democratic Republic of Congo (Kouerey Oliwina et al. 2020). Olive ridleys in Bioko are, on average, 5 cm larger than those nesting in Ghana, and leatherbacks there are, on average, 14 cm larger than those nesting in Ghana (Tomás et al. 2010). This size discrepancy may indicate that turtles nesting in Ghana are generally younger, possibly a result of intensive long-term human take in the fisheries or on the nesting beach (Bjorndal et al. 1985; Limpus et al. 2003).

Sea turtles are protected in Ghana under the Wildlife Conservation Regulation LI 685 (1971), but enforcement is limited due to inadequate resources and limited Wildlife Division (WD) offices along the coastline. The WD conducts occasional law enforcement patrols along nesting beaches near the offices, but those arrested are usually not fully prosecuted. To the best of our knowledge, no individual has ever gone to prison for violating sea turtle laws in Ghana.

The WD conducts education and awareness programs in schools and communities to improve knowledge about marine conservation. The WD uses elders and volunteer groups to involve the communities in developing sea turtle protection strategies. Some communities still respect a traditional story that considers it taboo to harm sea turtles (Alexander et al. 2017). This taboo provides protection for sea turtles in some regions, but this traditional reverence is quickly being lost to the younger generation, so they no longer observe this traditional taboo.

Sea turtles in Ghana, therefore, continue to face an increasing number of threats associated with fishing gear, direct take, and degradation of nesting habitat. Predation does not seem to be a significant problem on most sandy beaches in Ghana, although some beaches experience significant predation from feral dogs. Poaching is quickly eliminated on survey beaches because the presence of a survey team deters the poachers from attempting to take any turtles. Therefore, involving community members, especially fishers, as active participants in sea turtle management (i.e., nesting survey team) may be the most effective way of reducing anthropogenic sea turtle mortality in Ghana.

Although it appears sea turtle nesting is currently declining at this site in Ghana, historical data are not available to determine if this is a recent decline or something longer that may suggest a significant population loss across the region. Sea turtle recovery in Ghana will require a coordinated effort to survey beaches throughout the coastline to better document the current status and nesting seasons for each species. It will be important to determine hatching success as a way of quantifying reproductive success and the subsequent recruitment into the eastern Atlantic populations. Community education programs and improved enforcement of existing environmental protections regulations will be necessary to reduce mortality risks of nesting females and other turtles at sea.