Methods

The Nansha Archipelago (Spratly Islands) is composed of 102 coralline island, atolls, and shoals. Taipin Tao (10°22′N, 114°22′E) is located near the middle of the northern archipelago, which was 1365 m long and 360 m wide, with an area of 0.48 km². The island is forested with lowland tropical rainforest and reaches an average topographic height of 3.8 m (Fang and Li 1994). The sand particles are mainly shell debris and fine coral fragments. Coral reefs fringe the island and extend 100−200 m offshore. The monthly average air temperature ranges from 24.2°C to 32°C, with a yearly average of 28.1°C. The yearly total precipitation is more than 1800 mm (Naval Weather Center 1996, and this study).

The researchers were allowed to remain on the island for 5 days in 1995 (Cheng 2000a) and for 36 hours on each of the consecutive trips. Because of these limitations, the nesting ecology of sea turtles on the island was surveyed mainly by means of questionnaire interviews of the soldiers on the island in 1995 and by daily records of the patrol team, which walked the beaches around the island once every 3 hours from 1900 to 0500 hours and from the guards on the watch posts by the beaches each night (2000 to 2002). The beach was divided into 9 sections, based on the location of the watch posts where the patrollers started their field surveys every night (Fig. 1). The records include the date, species, site, emergence phase (crawling, digging, nesting, or descending), size (curved carapace length; CCL), health status, and fate of these turtles.

View Figure

• Download as Powerpoint
• Download Figure

To determine the sex ratio of the hatchlings on the island, 3 temperature loggers (Onset Optic StowAway Logger) were buried in the nesting beaches around the island at a depth of ∼70 cm in April 2002. The loggers were set to record sand temperatures every 1 hour 12 minutes. Thus, 22 sand temperature records were recorded each day, with a total of 7428 measurements over 372 days. Daily air temperature and precipitation were obtained from the Naval Weather Forecast Center, Taiwan, R.O.C.

To study postnesting movements of green turtles, I attached satellite transmitters in 2000, 2001, and 2003 (n =  2 each year). The Argos-linked satellite transmitters (ST-14; Telonics, Mesa, AZ) were attached following Balazs et al. (1996), later described in Cheng (2000b). The set-up of the parameters on transmitters and the criteria for data acceptance followed those of Cheng (2000b). Briefly, 7 different location classes (LC) are used by Argos. LC 3s are the most accurate location class and have an estimated accuracy of < 150 m. LC 2 is relatively less accurate, with an estimated accuracy range of ≥ 150 m to < 350 m, and LC 1 has an estimated accuracy of ≥ 350 m to < 1000 m. LC 0 is given an estimated accuracy of > 1000 m. LC A and B have no estimated location accuracy, and LC Z has an unknown accuracy. To ensure a reasonable level of integrity for our data, only LC B or better, which provide a reasonable estimation of the migration path, were accepted, i.e., when positioning was near LC 1, 2, or 3, and the distance between the 2 points reasonably reflected the swimming speed of a turtle; the location was not on land. Such positions were considered accurate and acceptable for the purpose of this study. Travel rate was calculated by dividing total travel distance (in km) on each trip by the total travel period (h). The end of migration was determined by the turtle stayed in the vicinity of one site over a month.

Results and Discussion

Both the questionnaire interviews in 1995 and the field daily records from 2000 to 2002 suggest that the nesting includes both green and hawksbill turtles (Eretmochelys imbricata). Whereas, soldiers interviewed in 1995 suggested that about 10% of the nesting females were hawksbills (Cheng 1996), the field daily records by coast guard personnel suggest that only 5% were hawksbills. Results of the patrol records showed that the CCL of the nesting green turtles ranged from 83 to 117 cm (mean = 98.5 ± 7 cm, n = 106). Turtles initially recorded as hawksbills ranged from 87 to 104 cm; however, these values exceed the size of local hawksbills (i.e., 66–86 cm) and are considered to be misidentified green turtles (Eckert 1993). It, therefore, is not possible to determine if species other than green turtles nest on Taipin Tao.

Of the 127 records, about 37% were in the nesting phase, 36% were in the crawling phase, 1% had nested, 4% were in the digging phase, 9% were trapped in the defense trenches, 2% were dead, and another 11% were either not recorded or observed from a distance. Thus, about 42% of the turtles were found engaged in nesting activities. Among the live turtles, 88% were healthy and 12% were wounded. Wounds were caused largely by the barbed-wire barricades as turtles crawled beneath them during the nesting attempts. To improve the nesting environment, the coast guard removed all the barricades in recent years.

Based on daily records from 2000 to 2002, approximately 106 nesting emergences occur per year; interviews suggest that approximately 55 emergences occurred in 1995. Daily field records from 2000 to 2002 indicate that the turtles nested all around the island (Fig. 1). The main nesting sites were on sections C and E to G. These sections were away from developed areas on the island (close to sections C to A and I), as has been found in other studies (Carr and Carr 1972; Wang and Cheng 1999; Cheng 2000a). Records also showed that nesting occurs year round (Fig. 2). This is true for many tropical sites, such as Sarawak (Chaloupka 2001), Antigua Island, Suriname (Glen and Mrosovsky 2004), Nacional Marino Las Baulas of Costa Rica (Reina et al. 2002). The peak season occurred from summer (July) to fall (November) and lowest from April to June.

View Figure

• Download as Powerpoint
• Download Figure

Research using satellite transmitters is a powerful tool for studying the migration of animals, such as the sea turtles (e.g., Hughes et al., 1998; Bentivegna 2002; Hays et al. 2002; Luschi et al. 2003; Craig et al. 2004). It provides not only the location of the resting/foraging sites of the turtle but also the migration behavior (e.g., Cheng 2000b; Polovina et al. 2000, 2004; Hays et al. 2002; Luschi et al. 2003). The research using satellite transmitters from 2000 to 2003 showed that the size of the turtles used in this study ranged from 91 to 103 cm CCL, 86 to 97.5 cm straight carapace length (Table 1). All turtles were tagged in April, and they stayed in the coastal waters of Taipin Tao from 1 day to 7 weeks before initiating their postnesting migrations. The turtles swam 9 to 40 days, with the travel distance that ranged from 327 to 1569 km. The travel rate ranged from 1.01 to 1.63 km h⁻¹. These values fall within the range of the other studies (e.g., Keinath and Musick 1993; Luschi et al. 1996; Cheng 2000b; Hays et al. 2001; Godley et al. 2004; Craig et al. 2004). The migration behaviors were found to be mostly unaffected by ocean currents in the region (Cai Y.R., pers. comm.). Most turtles swam directly to their coastal foraging sites, although 2 turtles showed biphasal migration in which the turtles migrated along the coast for a portion of their movements before moving into deep waters of offshore Taipin Tao (i.e., NS1-00 and NS2-00; Fig. 3). These results were similar to the studies on Wan-an Island (Cheng 2000b) and consistent with movements of turtles reviewed by Luschi et al. (2003). Despite the longer migration distance, green turtles nesting at Ascension Island also moved directly to their foraging grounds off Brazil after nesting (Luschi et al. 2003). Green turtles feed on seagrass and macroalgae, however, these food sources are not available in the open ocean, which may result in the direct migrations to the coastal foraging sites. The end point of the migrations included Palawan Island, Philippines; northeast coast of Luzon, Philippines; East Sabah, Malaysia facing the Sulu Sea; and north Borneo, near Labuan, Malaysia (Fig. 3). These results suggest that southern Palawan Island, Philippines, is one of the foraging sites for green turtles nesting at Taipin Tao. This region is also the foraging site for green turtles nesting in Malaysia (Liew et al. 1995).

Table 1. Size, tagged and release date, postnesting migration distance, traveling duration, and travel rate of the green turtles tagged with the platform terminal transmitter attached to green turtles nesting at Taipin Tao, Nansha Archipelago, between 2000 and 2003.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Conservation Implications

The South China Sea likely hosts the majority of sea turtle population in China. However, political instability, occupation of the land by the military, intentional capture of nesting and foraging areas, habitat loss or degradation, incidental captures in marine fisheries gear; marine pollution; and illegal, uncontrolled, and unreported fishing are well-known and serious threats to the marine biodiversity in the South China Sea (Liang et al. 1990; Nishemura 1990; Cheng 2000a). These, in turn, exert an extremely detrimental impact upon the already endangered sea turtle populations. Nesting islands, such as Taipin Tao, therefore, are crucial to the survival of the sea turtle population in the region.

The coast guard on the island should be encouraged to actively protect sea turtle populations, both on the land and in the sea. In addition, more ecological studies on sea turtle nesting should be carried out, such as nesting frequency studies and satellite telemetry, to provide valuable information on the population dynamics and dispersal of sea turtles from Taipin Tao. Research using satellite transmitters clearly demonstrated that the green turtles nesting at Taipin Tao are a shared natural resource among the countries in the region. The overlapping of the foraging sites in the coastal waters of southern Palawan Island with other nesting populations further suggests that this area is a key foraging site for green turtles from the South China Sea. Furthermore, the widespread dispersal over the southwest region of the South China Sea suggests that, as at Wan-an Island in the Penghu Archipelago of Taiwan (Cheng 2000b), the conservation of the marine habitat is as important to the survival of sea turtle populations as is nesting beach protection. Thus, a regional program and strategy under the umbrella of international organizations, such as the Association of Southeast Asian Nations and the Indian Ocean and Southeast Asian Memorandum of Understanding) is needed for Taipin Tao (MTSG 1995). For example, in response to the construction of a runway on the island by the Defense Department for the national security, the Kaoshiung City government, which exercises jurisdiction over Taipin Tao, has announced that the nesting beaches and surrounding 12 nautical miles of Taipin Tao is a “Green Sea Turtle Reproductive Site” (effective on 3 March 2007).