Green turtles (Chelonia mydas) occur in all temperate and tropical waters throughout the world (Seminoff 2004). However, their current populations are a fraction of their historical levels and are under continual stress from a variety of human-induced activities (Awbrey et al. 1984; Seminoff 2004; Koch et al. 2007). The US Endangered Species Act lists the species as threatened globally, with the Florida and Pacific Mexico populations listed as endangered (Seminoff et al. 2007). This species is also listed as an endangered species in Norma Oficial Mexicana 059 (SEMARNAT 2001) and has been listed as endangered by the International Union for Conservation of Nature and Natural Resources since 1982 (Seminoff 2004). Two conservation issues are the loss of nesting beaches with no human harassment or development and the loss of populations due to indiscriminate hunting and egg collection (Seminoff et al. 2003). Remote oceanic islands that have little direct human impact provide key breeding sites to ensure the survival of this species.

Brattstrom (1955) first reported green turtles nesting on Isla Clarión in the Mexican Archipelago Revillagigedo. Awbrey et al. (1984) and Sarti et al. (2007) emphasized the importance of the Revillagigedo Archipelago for green turtles (known locally as “black turtle”) because of their use of the islands for foraging and reproduction. The archipelago is also used by hawksbill turtle (Eretmochelys imbricata) juveniles, and occasionally olive ridley (Lepidochelys olivacea) and leatherback (Dermochelys coriacea) turtles for foraging, but not nesting (Sarti et al. 2007). However, Sarti et al. (2007) did not provide any quantitative data on their green turtle surveys, and Awbrey et al. (1984) reported only a single count of nest pits.

Here we document our observations of nesting crawls and hatching events by green turtles on the beaches of Isla Clarión. We report the number of beach emergences by adult females and the number of groups of hatchlings on this remote island during the second half of November 2008.

Study Area

Isla Clarión is the most westerly of the four volcanic islands of the Revillagigedo Archipelago, Mexico, in the eastern Pacific Ocean. The island is located at 114°45′W and 18°21′N, 1100 km west of Manzanillo, Mexico. The Archipelago is listed as a Ramsar (2005) reserve and a Biosphere Reserve by the Mexican government (CONANP-SEMARNAT 2004). The waters around Isla Clarión are listed as a Mexican Marine Reserve.

Isla Clarión has 3 stretches of sandy beach, all located on the south coast. Sulphur Bay Beach and East Beach are separated by the volcanic headland, Farallón de la Bandera. In addition, a small beach lies at the west end of Sulphur Bay and the boat ramp (West Beach). All the beaches were 5–10 m wide (water edge to back of beach), ending uphill in a shallow dune system. The lengths of the beaches were measured by a Garmin GPS unit.

Sulphur Bay Beach is curved, approximately 950 m in length, and is mostly coarse sand with small patches of large coral. The beach terminates in black volcanic rocks at both ends. Toward the west end of this beach is 100 m of coral reef, which is submerged at high tide but is partially exposed at low tide. Low sand dunes colonized by the forb Hipomoea pre-caprae occur continuously along the full length of this beach; turtle body pits (from nesting events) form most of the relief on these dunes.

East Beach consists of mostly large coral and the center section has a very steep gradient, with 50-cm vertical faces interspersed with 4-m vertical heaps of coral that appear inaccessible and unsuitable for nest activity. However, the east and west ends of this beach have 300-m stretches of sandy beach and a sandy ridge connecting them that appears suitable for green turtle nesting.

West Beach is a relatively short (50-m) beach at the extreme west end of Sulphur Bay, near the boat ramp. The beach is similar in structure but slightly steeper than Sulphur Bay Beach.

Methods

From 19 November 2008 to 4 December 2008 we conducted a daily census of the turtle tracks on Sulphur Bay Beach to quantify emergences by female green turtles. Sunrise during this interval occurred between 7:47 AM and 7:56 AM (Central Standard Time [CST]) according to the US Navy web site (http://www.usno.navy.mil/USNO/astronomical-applications). Each morning between 7:00 AM and 8:00 AM CST, we walked the Sulphur Bay Beach and counted the number of female turtle tracks that emerged from the water and climbed over the top of the dune; however, we did not count tracks that emerged and returned to the sea without making it to the top of the beach. Although we did not attempt to confirm that each emergence resulted in a successful nesting event, we did match each emergence crawl with a return (to the water) crawl. We determined which tracks were from the previous night and which were at least 24 hours old using a track-marking system. Older tracks were not counted. Females observed on the beach during our dawn patrols were encountered as a result of our following their emergence crawl to the upper dune area of the beach; in these cases we did not match emergence crawls with return crawls.

We also counted the concentrated groups of tracks from hatchlings that were new each day and headed down slope. We assumed tracks in the same general area originated from one nest and, if dense sets appeared relatively close together, we confirmed that they came from the same nest by pinpointing the starting points for each track set. Solitary hatchling tracks were not considered a hatching event; we assumed these were from hatchlings that had crawled parallel to the water's edge before leaving the dunes and, thus, part of a nearby hatch.

We determined that the tracks were new each day by walking on the old tracks each morning and only included tracks of females and hatchlings that were on top of our footprints on each morning's count. In the second week of our surveys high tides in the afternoon cleaned the beach of tracks for the next morning's count, making it easy to confirm tracks were only from the previous night.

Results

We recorded 115 complete tracks of females during the 16 days that we conducted surveys on Sulphur Bay Beach (Table 1). The number of daily emergences increased each day from 19 November to its peak on 23 November (n  =  15 emergences), 4 days before the new moon (Fig. 1). From 24 November to 3 December, there were 3–9 emergences each night. From 20 November to 25 November, we observed adult green turtles still in the dunes or on the beach at dawn on 12 occasions; in all cases, turtles departed the beach shortly after dawn. After 25 November, no adult green turtles were seen on the beach or dunes during our dawn patrols.

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Summary of adult green turtle emergences and green turtle hatching events observed on Isla Clarión between 19 November 2008 and 4 December 2008.

View Fullscreen

We recorded 219 green turtle hatching events on Sulphur Bay Beach during 14 days of searching for such events (Table 1). The number of hatching events per night varied from 4 to 27 with a mean of 15.6 (Fig. 2). Among hatching events, the number of discernable hatchling tracks varied greatly.

View Figure

• Download as Powerpoint
• Download Figure

On 22 November we walked the length of East Beach and saw no evidence of turtles climbing the middle, coral section of beach. On the east end of this beach we counted 12 tracks of females and 4 sets of hatchling tracks; on the west end of this beach we counted 6 adult female tracks but no hatchling tracks. We believe that these counts included tracks of females from the prior 3 nights. Thus, we counted 18 tracks of females over 600 m or 1 track per 100 m per day. During the morning (22 November), we counted 29 tracks of females on the 950 m of Sulphur Bay Beach, or 1 track per 100 m per day. As at West Beach, these tracks were believed to represent nesting activity from the prior 3 nights.

We used all these numbers to estimate the total number of nesting crawls by female turtles that emerged to lay eggs during these 2 weeks. Because we recorded 115 nesting crawls between 19 November and 4 December on the Sulphur Bay Beach or 12 crawls per 100 m, we extrapolate that the 600 m of the East Beach supported an additional 72 crawls during our 16-day visit for an island total of 187 crawls. Our assumption that the number of crawls on our 1 survey of the East Beach was representative of the turtles' use of that beach throughout the 2 weeks was supported by the identical number of crawls per day per 100 m on both beaches. On 23 November we walked the West Beach at the small boat ramp and recorded 4 fresh tracks, and on 30 November only 1 new track was obvious, for an average of 2.5 per day or 40 additional females. Therefore, we estimate that about 227 crawls of female turtles occurred on the 3 beaches during our 2-week stay.

We estimated the total number of hatches for all beaches using the daily surveys of the West Beach and our single count on the East Beach on 22 November. We recorded only 4 sets of tracks of hatchlings on the two parts (total 600 m) of the East Beach, compared to 15 on the 950-m beach of the West Beach. These numbers give rates of 0.6 tracks per 100 m versus 1.6 tracks per 100 m per day on the East and Sulphur Bay beaches, respectively. Because the long East Beach may have only 40% as many hatches per 100 m and is 0.63 as long, we can extrapolate the 219 hatches in 14 days on Sulphur Bay Beach to 219 × 0.4 × 0.63  =  55 hatches on the East Beach, for a total of 274 hatches on both beaches for 2 weeks. Thus, more females were laying eggs previously, as represented by the number of hatches that emerged during our 2-week stay on the island. We did not find any hatchling tracks on the short West Beach, so did not add any hatches for this section of beach.

Discussion

Obviously we did not see most of the turtles to confirm that all were green turtles. However, our species identification was based on the 12 females that we saw near dawn on the first week of our survey, on the 4 sets of hatchlings that we encountered, and the fact that none of the tracks were obviously different from all the others (which might have suggested any other species was nesting on the island while we were there). We did not determine if the turtles were of the “black” or “green” variety, which Juarez-Ceron et al. (2003) described as both being present on the islands.

Our population estimate could be imprecise for several reasons. Pritchard (1971) noted that female green turtles made more than 1 visit to beaches over a few days. Thus, our count of female emergences could include repeat visits, and the actual number of females laying could be fewer than the number represented by our track counts. Also, we surveyed the East Beach once and the West Beach twice, which may not have been representative of the 2 weeks. And the 2 weeks of our surveys are not necessarily representative of the nesting season.

We found few data available to make comparisons with our counts. Brattstrom (1982) stated that green turtles nested on only 1 beach on Sulphur Bay. However, he must have not visited the East Beach, which is suitable for nesting turtles. In 3 days in October 1976, Awbrey et al. (1984) counted 80 body pits that they estimated were dug in the previous year on Isla Clarión and determined a density of 0.027/m². We suspect this total is less than the annual number of nests because females will dig over other pits, especially those that have hatched. The nesting pits are the major relief on the dunes, and no area of dunes was not dug up. Awbrey et al. (1984) counted 30 fresh female tracks in 3 days on Sulphur Bay Beach, comparable to the number recorded in our survey. Juarez-Ceron et al. (2003) counted a mean of 79 nests (65–105) during an unreported number of visits from 1999 to 2001 on all 3 Isla Clarión beaches. Our survey indicated that many more females were nesting on Isla Clarión than was estimated previously. Our total number of hatching events in 2 weeks is greater that the totals by both these previous surveys.

Green turtles have an extended breeding season on Isla Clarión. Brattstrom (1982) recorded fresh tracks as early as 25 March (1953), on 7 May (1955), and as late as 9 November (1971). Awbrey et al. (1984) recorded nesting on 9–11 October (1976). Juarez-Ceron et al. (2003) stated that nesting was year-round with peak activity during October and November. Sarti et al. (2007) stated they found turtles nesting from April to December. Our survey results confirm the later laying dates. Green turtles could nest here year-round as predicted by Hendrickson (1958) for equatorial locations with warm water year-round, as occurs at Isla Clarión (Sharp 1978).

Most turtle hatching occurred at night as expected (Mrosovsky 1968). However, we witnessed predation of hatchling turtles almost daily by Clarión ravens (Corvus corax clarionensis). Groups of ravens congregated daily at turtle hatching sites searching for late hatchlings at dawn and dusk, as also reported by Awbrey et al. (1984). They walked over nest areas digging through the sand with their bills until they successfully found a hatchling. At one point we witnessed a raven using a piece of dried sponge-like material as a scraper to more effectively move the sand. Although small numbers of frigatebirds (Fregata magnificens) were seen over the island, we did not record their taking of young turtles, as they do very quickly in Galapagos (Pritchard 1971). One small octopus (Octopoda sp.) that was harvested by a Mexican marine expelled a baby turtle.

On the morning of 28 November 2008, we noted fresh human footprints where a female turtle had gone to lay eggs. At breakfast that morning approximately 60 turtle eggs were boiling in a large pan of water in the garrison kitchen. Obviously, turtle eggs should not be harvested, and turtle nests should not be disturbed, in accordance with Mexican law (Diario Oficial de la Federación 1990). We believe this was an isolated incident during our visit. The Mexican Navy has developed a conservation plan for the Revillagigedo Archipelago (CONANP-SEMARNAT 2004), and we encourage its implementation, including education of the marines stationed on the island about the conservation issues.

Seminoff (2004) cautioned about making extrapolations based on short-term monitoring. We do not know how many turtles nest on Clarión each year. The number of sets of hatchling turtle tracks was greater than the number of female tracks during our stay, indicating that more females had used the beaches previously. Juarez-Cerron et al. (2003) found nest success was high at an average of 89.7% for Isla Socorro, which would indicate the number of females using the beach in October would have been higher than in December. A combination of the two types of tracks indicated that > 500 nests may have been in various stages of development on the beaches in two 2-week periods (nestling hatches and track counts combined). With at least an 8.5-month nesting season, a considerable number of female turtles could nest on Isla Clarión each year. Although this number is small compared to the numbers nesting on mainland Michoacan (around 6000 nests annually), this population has been found to be genetically distinct, and Isla Clarión is the most important nesting site in the Archipelago (Sarti et al. 2007). The other beaches suitable for turtle nesting are 5 small beaches on the north side of Isla Socorro (Brattstrom 1982). Here, Juarez-Ceron et al. (2003) found an average of 47 nests over 3 years (1999–2001), considerably fewer than on Isla Clarión. Lorvelec et al. (2009) confirmed that green turtles do not nest on Clipperton Island. Thus, the nesting populations on Galapagos Islands (Pritchard 1971) and these on the Revillagigedos are the only far offshore islands where this species breeds in the eastern Pacific Ocean.

Our survey suggests Isla Clarión may support a larger number of green turtles than previously recorded and further counts are warranted over a more extended time period. Efforts should be made to preserve the isolation of the Isla Clarión beaches to protect this breeding site. Awbrey et al. (1984) stated that turtles on the near-shore islands of Baja and Mexico were subject to human-caused mortality. Thus, these offshore island populations may be critical to the survival of this species in the eastern Pacific Ocean.

Furthermore, Isla Clarión would be an excellent site to conduct more detailed studies of this species at a site isolated from direct human impacts. Such studies could be compared to the status of turtles at sites with human impacts (Seminoff et al. 2003). Telemetry studies would show the importance of Isla Clarión as a source rookery for coastal areas in Mexico, as has been shown in Galapagos Islands (Seminoff et al. 2008).