Laparoscopy of Nesting Hawksbill Turtles, Eretmochelys imbricata, at Milman Island, Northern Great Barrier Reef, Australia
ABSTRACT
Laparoscopies were performed on 225 hawksbill turtles (Eretmochelys imbricata) following nesting emergences on Milman Island, northern Great Barrier Reef, Australia during the 1992–1995 summer seasons. New recruits (first time breeders) made up between 9.8% and 14.3% of the nesting population across 4 summer seasons. Remigrants were not significantly larger or heavier than recruits, indicating that females nesting for the first time do not necessarily recruit to a nesting population at the smallest sizes. Further investigations are required to determine if the summer recruitment rate is reflective of the entire nesting population, which nests year-round.
Tagging studies are valuable tools for determining size of nesting marine turtle populations (Hendrickson 1958; Carr and Carr 1969; Limpus 1985) but they require long-term (e.g., decades) programs to ascertain reproductive history of newly recruited nesting females (e.g., first time breeders). Although these studies have been used to identify new recruits to a population and determine remigration intervals for a species, they have provided little information on previous breeding history. Laparoscopic examination of postnesting crocodiles and turtles has proven a useful tool in identifying recruits as well as determining sex and age class composition of marine turtles at foraging grounds (Wood et al. 1983; Limpus 1985, 1992; Limpus et al. 1994). Determining the proportion of a population that are recruits can give an indication of the population stability. For example, Limpus et al. (2003) noted a relatively low proportion of green turtles nesting at Raine Island are new recruits, possibly indicating a loss of female turtles reaching reproductive maturity. This study summarizes laparoscopic results related to age class composition of a population of hawksbill turtles (Eretmochelys imbricata) nesting during the Austral summer months on Milman Island, northern Great Barrier Reef, Australia.
Methods
Laparoscopies were performed on hawksbill turtles that completed successful nesting emergences at Milman Island, northern Great Barrier Reef, Australia during portions of the 1992, 1993, 1994, and 1995 summer nesting seasons (Table 1). These periods encompassed peak and post-peak nesting activity (Dobbs et al. 1999). Animals were randomly selected for laparoscopy after laying eggs, thus providing an indication of recruitment rate during each study interval. Following protocols described by Limpus et al. (1983), most undergoing laparoscopic examination (i.e., “lapped” turtles) were measured (curved carapace length; CCL) and weighed (Table 2).
After laying eggs, turtles were secured to a wheelbarrow with their plastron facing up. The laparoscopic examination was performed by JDM and ovaries assessed following established criteria (Limpus 1985, 1992; Limpus et al. 2003). Limpus and Limpus (2003) provide a key for examining marine turtle gonads via laparoscopy. Turtles were examined for the presence or absence of corpora albicantia, which are ovarian scars from ovulations in previous years (Limpus et al. 2003). Lapped turtles were categorized as 1) “remigrant” if the turtle was either tagged in a previous nesting season and was returning to nest at a known remigration interval, or was tagged in the current nesting season and laparoscopy revealed corpora albicantia, indicating the turtle had bred in a previous nesting season; or 2) “new recruit” if the turtle was tagged in the current nesting season and laparoscopy revealed no corpora albicantia, indicating the turtle had not bred before the current nesting season. Results were analyzed following statistical criteria in Zar (1974).
Results
Laparoscopies were performed on 225 hawksbill turtles (Table 1), 25% of 895 turtles recorded nesting during the peak of the 1992−1995 summer seasons (Dobbs et al. 1999). Almost half of the turtles nesting in 1992 (48%) and 1995 (40%) were lapped, with less than 20% in 1993 (11%) and 1994 (17%). The difference in proportion of turtles lapped relates to the amount of time JDM was available on the island to conduct the procedure.
Twenty-four (10.7%) of the lapped turtles were new recruits with no corpora albicantia while 31 others (13.8%) were remigrants recorded at Milman Island during at least 1 previous nesting season (Table 1). Recruitment rate (proportion of new recruits from all lapped turtles) to the nesting population ranged from 9.8% to 14.3% in each sampling period (mean = 11.1 ± 2.1%) (Table 1).
Average CCL and post-ovipositional weight of lapped turtles were not significantly different between remigrants and new recruits (t = 4.42, p > 0.03; t = 2.52, p > 0.11, respectively; Table 2). Two new recruits and 6 remigrants did not have CCL recorded, whereas 2 new recruits and 10 remigrants were not weighed.
Corpora albicantia diameters ranged from 1.0 to 6.0 mm in all remigrants (Fig. 1); however, remigrants with the shortest remigration interval (2 years) had the largest average corpora albicantia diameter (3.4 mm, Fig. 2). One turtle, T55440, encountered nesting in 3 seasons at 2-year intervals (1991, 1993, 1995) had 2 size-classes of corpora albicantia, 2.0 and 4.0−5.0 mm.
Citation: Chelonian Conservation and Biology 6, 2; 10.2744/1071-8443(2007)6[270:LONHTE]2.0.CO;2
Citation: Chelonian Conservation and Biology 6, 2; 10.2744/1071-8443(2007)6[270:LONHTE]2.0.CO;2
Approximately 34% (77 of 225) of lapped turtles were seen in subsequent nesting attempts in the same season. Based on the number of mature follicles recorded at the time of laparoscopy, at least 79 turtles were recorded as having 50 to 80 or more mature follicles indicating they had at least one more clutch to lay or were just beginning nesting for the season. Twenty turtles were likely to have completed nesting for the season at the time of laparoscopy because they had no (n = 5), < 10 (n = 1), few (n = 11), or < 20 (n = 3) mature follicles. The remaining 126 turtles did not have any recorded information on abundance of mature follicles when they were lapped so it is not known if they could have been expected to return to nest again during the season.
Turtles returning to nest after laparoscopy had lost their stitches and exhibited little evidence of the surgical procedure (K.A. Dobbs, pers. obs.). Renesting intervals for lapped turtles, calculated as both days to re-emerge (15.4 ± 2.5 days, range = 12–25, n = 73), and days to renest (15.5 ± 2.5 days, range = 12–25, n = 71) were significantly longer than those recorded for unlapped cohorts for days to re-emerge (mean = 14.4 ± 1.6 days, range = 10–21, n = 1285, t = 22.77, p < 0.001), and for days to renest (mean = 14.7 ± 1.8, range = 10–25, t = 13.9, p < 0.001).
Injuries resulting from the laparoscopy procedure were recorded in 27 turtles (12%). Of these, punctures of the lung, which can be more serious if not managed correctly, were most frequent (n = 24); followed by yolk (n = 2) and ovarian blood vessel (n = 1) punctures. During surgery it was noted that the lung of hawksbill turtles often extended around the gastrointestinal track toward the surgical entry area when the turtle was inverted, making it more vulnerable to puncture than in green and loggerhead turtles. Adjustments made to the surgical process to reduce the incidence of puncture included changing the location and angle of entry in the inguinal pocket as well as controlling the depth of entry. Since implementing these processes, the incidence of lung puncture in hawksbill turtles has diminished.
Of the turtles with lung punctures, 6 were expected to nest again, 4 were finished for the nesting season, and 14 did not have any information about whether they were expected to nest again in the season, although one of these turtles was recorded nesting again in the same season. Five of the turtles with lung punctures have been resighted in subsequent nesting seasons.
Discussion
Recruitment rate fluctuated from 9.8% to 14.3% among field seasons in this study. Hawksbill turtles nest year-round on Milman Island (Dobbs et al. 1999; Limpus 1980; Miller 1994), and laparoscopies of hawksbill turtles nesting in seasons other than summer have not been conducted. The laparoscopic assessments reported herein were made at different times of the summer peak-nesting season, and further information is required to determine whether recruitment rate is uniform throughout the year and between nesting seasons.
Average annual recruitment rate reported here (11.1%) is similar to that for hawksbills nesting on Antigua (< 20%; Hoyle and Richardson 1993), flatback turtles (Natator depressus) nesting at eastern Australian rookeries (10%−20%; Limpus et al. 2002) and loggerhead turtles (Caretta caretta) nesting on Heron Reef, Australia (13.1%–17.4%; Limpus 1985). However, it was much less than the reported 27.2% of nesting flatback turtles on Crab Island (Limpus et al. 1984) and 20%–30% for loggerhead turtles nesting in Tongaland (Hughes 1974). Witzell's (1983) synopsis of hawksbill turtle biology does not provide information regarding recruitment rates in nesting populations.
Remigrant hawksbill turtles were not significantly larger or heavier than new recruits on Milman Island, indicating first time nesters do not necessarily recruit to a nesting population at the smallest sizes. This is consistent with results from the southern Great Barrier Reef green turtle population (Chaloupka et al. 2004) but contrary to that for the northern Great Barrier Reef green turtle population (Limpus et al. 2003) and for loggerhead (Limpus 1991) and flatback (Parmenter and Limpus 1995) turtles in the Great Barrier Reef. There was some indication that new recruits have a smaller CCL than remigrants (Table 2); however, this characteristic is quite variable and further sampling is needed to determine whether there is a significant difference in carapace length. The reasons hawksbill turtles do not recruit to the Milman Island nesting population at the smallest sizes are unknown, but may be an indication of the quality of the foraging habitat in which they live or maturation may not be a function of size alone. Increasing the sample sizes of hawksbill turtles laparoscopied at Milman Island and knowledge of the home foraging areas of this nesting population may assist with understanding this reproductive parameter.
The rate at which ovarian scars regress from corpora lutea to corpora albicantia provides an indication of how long it has been since the animal nested (Wood et al. 1983; Limpus and Reed 1985a, 1985b; Limpus 1992), and may be used to estimate remigration intervals. Hawksbill turtles encountered at 4-year intervals had the largest proportion of corpora albicantia diameters in the 1.0−2.0 mm range, and the smallest overall average corpora albicantia diameter (Fig. 2). Collecting information on corpora albicantia diameter from turtles hunted or found slaughtered at markets could provide an additional means for assessing impacts of such harvests on the various turtle populations being harvested (e.g., are harvests targeting breeding females, first time breeders?).
Conducting a laparoscopy appeared to increase the average renesting interval of hawksbill turtles by 1 day. Hawksbill turtles nesting at Milman Island are known to nest within the same season at nearby islands (Bell et al. 1999). It is not known if any of the lapped turtles nested at these locations. Also, because of the timing of laparoscopy work at Milman Island within the context of an overall nesting survey, it was not always possible to remain at the island to record whether lapped turtles returned to nest and if so, at what interval.
Trauma associated with laparoscopic procedure (e.g., lung punctures) needs to be identified and considered when selecting research methods and assessing their impacts on animals. Laparoscopy provides a relatively safe and valid method for determining recruitment rates as long as adequate training in this procedure has occurred and adjustments in the surgical process are made to account for species specific variation in internal anatomy; however, other noninvasive techniques, including ultrasound imaging (Rostal et al. 1990) might prove useful as technology improves. The resighting of turtles with punctured lungs in subsequent nesting seasons is encouraging for those animals that were not resighted during the season they were lapped. However, the lack of a resighting needs to be interpreted with caution. The nesting beach monitoring at Milman Island since February 2005 has generally been conducted for between 2 and 4 weeks, rather than the 8 to 12 weeks during this study, so there has been less of a chance to resight these turtles.
Determining past reproductive history of nesting marine turtles is important when assessing trends in the population. Populations with a high recruitment rate may have experienced a loss of adult females while those with low rates may have experienced a loss of new recruits from the population. Just as important, however, is locating foraging areas used by turtles and quantifying their maturity ratio in those areas. This will help determine the proportion of a population nesting within a season and possibly highlight issues associated with the foraging ground that may delay or inhibit turtles from maximizing their reproductive potential (e.g., human-related impacts on forage material).
Diameter of corpora albicantia in remigrant hawksbill turtles, Eretmochelys imbricata, nesting on Milman Island, Australia, during summers 1992–1995.
Remigration interval (years) by corpora albicantia diameter (mm) of remigrant hawksbill turtles, Eretmochleys imbricata, nesting on Milman Island, Australia, during summer 1994 and 1995.
