This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights Author's personal copy Journal of Experimental Marine Biology and Ecology 445 (2013) 140–147 Contents lists available at SciVerse ScienceDirect Journal of Experimental Marine Biology and Ecology journal homepage: www.elsevier.com/locate/jembe Estimating the sex ratio of green sea turtles (Chelonia mydas) in Taiwan by the nest temperature and histological methods Rowena King a,⁎, Wan-Hwa Cheng b, Cheng-Tsung Tseng c, Hochang Chen c, I-Jiunn Cheng c,⁎ a St. Vincent Girls' High School, P.O. Box 1057, Kingstown, Saint Vincent and The Grenadines b Department of Biological Science, Central Florida University, Orlando, FL 32816, USA c Institute of Marine Biology, National Taiwan Ocean University, Keelung 202-24, Taiwan, ROC article info abstract Article history: Sex ratio is an important population characteristic. It is especially important for endangered species with Received 23 July 2012 temperature-dependent sex determination, such as sea turtles. As turtle sex ratio at hatching is primarily Received in revised form 28 March 2013 influenced by the environmental temperature, the ability to estimate sex ratio is crucial to the success of conser- Accepted 29 March 2013 vation measures. Green turtles (Chelonia mydas) are the only sea turtles that nest in Taiwan. Despite much region- Available online xxxx al research on the ecology of this species, the sex ratio has not yet been determined in Taiwan. We combined measures of nest temperature during the thermally sensitive period with a histological method to estimate the Keywords: Green sea turtle hatchling sex ratio from three nesting islands in Taiwan from 2010 to 2011. We determined that the sex ratios Hatchling sex ratio for green turtle populations in Taiwan were female biased during the study period. In line with that overall Histological method trend, Wan-an Island with drier and hotter weather during the nesting season, produced more female hatchlings Nest temperature than the other main nesting island—Lanyu Island. Results of this study stress that differences in sex ratio are im- Nesting islands portant to consider when developing conservation strategies, even among closely located sites such as Wan-an and Lanyu Islands. This is the first study of the hatchling sex ratio of green sea turtles in the East Asian region. © 2013 Elsevier B.V. All rights reserved. 1. Introduction For immature turtles, the lack of morphological sex characteristics has made distinguishing the sexes difficult. (Larios, 1999; Wibbels, It is important to assess the sex ratios of endangered species with 2003). The sex of juvenile animals can be determined either by lapa- temperature-dependent sex determination, such as sea turtles, so roscopic observation of gonads (Limpus and Reed, 1985; Wibbels, that the conservation implications of skewed sex ratios can be consid- 1999, 2003) or indirectly by radioimmunoassay (RIA) of testosterone ered. Recent reviews of global climate change (e.g., IPCC, 2007) have in the blood serum (Owens et al., 1978; Wibbels, 1999). For hatch- demonstrated the impact of temperature on organisms worldwide. lings, two methods are currently employed: direct and indirect Because temperature is rising globally, it is important to determine (Wibbels, 2003). Direct methods involve sacrificing hatchlings or the effects of temperature and other factors on the sex ratios of sea using dead hatchlings found in the nest. Their sex can be determined turtle populations, and to learn whether their environmental sex de- by careful gross morphological examination of the gonad exterior and termination operates similarly among species and populations global- accessory duct characters (Wyneken et al., 2007) or by histological ly (e.g. Hawkes et al., 2009; Hays et al., 2010; Houghton et al., 2007; examination of the gonad (e.g. Ceriani and Wyneken, 2008; Newson et al., 2007; Poloczanska et al., 2009). Sea turtles are endan- Wibbels, 2003). Of these direct two methods, the latter is considered gered species with sex ratios, at least at egg hatching, determined by more accurate (Mrosovsky and Benabib, 1990; Mrosovsky and nest temperature. Information on the sex ratio at every life stage is Godfrey, 1995). It incurs fewer sex-identification errors than the crucial to conservation measures intended to protect these species. gross morphology method (Mrosovsky and Benabib, 1990). Indirect Several methods have been used to determine the sex of sea turtles. methods use pivotal temperature, sand temperature or incubation For adult turtles, sex identification is not difficult, because male duration to predict the hatchling sex ratios from the nesting beach turtles have a large and muscular prehensile tail extending well be- (e.g. Broderick et al., 2000; Godfrey et al., 1999; Godley et al., 2001; yond the carapace. The tails of females are short and, at most, project Marcovaldi et al., 1997; Mrosovsky et al., 1999, 2009). That reduces only slightly beyond the edge of the marginal scutes (Wibbels, 1999). the tedious work of sexing individual hatchlings. However, it also re- quires sacrifice of large numbers of hatchlings to obtain the pivotal temperature (PT) and Transitional Range of Temperature (TRT) ⁎ Corresponding author. Tel.: +886 2 24622192x5303; fax: +886 2 24628974. from a specific beach in the preliminary work (Wibbels, 2003). De- E-mail addresses: [email protected] (R. King), [email protected] (W.-H. Cheng), [email protected] (C.-T. Tseng), [email protected] (H. Chen), spite morphological differences it is difficult to estimate the sex [email protected] (I.-J. Cheng). ratio of adults and juveniles in any given sea turtle population. For 0022-0981/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jembe.2013.03.016 Author's personal copy R. King et al. / Journal of Experimental Marine Biology and Ecology 445 (2013) 140–147 141 the adults, only the females emerge to nest, and they do not nest 2. Material and methods every year. Despite this logistical problem, it is possible to estimate sex ratios of breeding adults from fishery by-catches and direct under- Two methods were used to determine hatchling sex ratios in this water observations. Juveniles are difficult to locate, and if found, it is study: the histology of dead hatchlings and the nest temperature uncertain what proportion of the population they represent (Wibbels, during the sex-determination period. The endangered status of green 2003). Sex identification of hatchlings is, therefore, the most reliable turtles in Taiwan and the low number of turtles nesting in Taiwan method to estimate the early-life sex ratio of a population. prevented us from sacrificing whole clutches as has been done in Several factors have been found to influence the sex ratio of hatch- some other studies (e.g. Godfrey et al., 1996; Mrosovsky, 1988). In addi- lings. They include latitude, season, shading such as by nearshore forest, tion, logistic limitations (e.g. insufficient manpower in the field, poor sand color, precipitation, nest depth and anthropogenic events like beach weather conditions, etc.) prevented us from extensively sampling the nourishment (adding sand) and adjacent housing projects (Broderick et clutches. To amend this shortfall, we recorded the clutch temperatures al., 2000; Godfrey et al., 1996; Hawkes et al., 2009; Hays et al., 1999; with loggers and estimated the clutch temperature from the sand tem- Houghton et al., 2007; Kamel and Mrosovsky, 2006; Kaska et al., 2006; perature near clutches for which loggers were not available. The sex Morreale et al., 1982; Mrosovsky, 1988; Mrosovsky et al., 1984; ratios of the green turtle hatchlings from clutches not examined directly Pavenport, 1997; Poloczanska et al., 2009; Wibbels, 2003). These factors can be approximated from the incubation temperatures. are all related to in situ sand temperature. Temperature influences the sex of the hatchling through the sex determination cascade, by activating 2.1. Field data collections genes encoding enzymes for steroid synthesis and for enzymes such as aramatase (Godfrey et al., 2003; Hulin et al., 2009; Merchant-Larios et Data were collected from Lanyu and Wan-an during the nesting al., 1997; Pavenport, 1997; Wibbels, 2003). This will result in the varia- seasons (June to October) of 2010 and 2011, with the addition of tion of sex ratios on both the spatial (different nesting beaches) and LiuChiu Island in 2011. Beaches were patrolled for nesting females temporal (yearly) scales. Therefore, it is important to determine the each night and early morning. Once a female had finished camouflaging sex ratios of turtles from different nesting beaches in order to conduct her nest, eggs were excavated within 4 h of oviposition and the clutch proper conservation measures. size (total number of eggs laid) was determined. Morning patrols Five species of sea turtle live near Taiwan, namely green turtle were conducted to make sure that no nest had been missed the previ- (Chelonia mydas), loggerhead turtle (Caretta caretta), hawksbill turtle ous night, and patrols were also conducted on the earliest expected (Eretmochelys imbricata), olive Ridley turtle (Lepidochelys olivacea) hatchling emergence date, so that the emerging hatchlings could be and leatherback turtle (Dermochelys coriacea). Among them, only recorded. Nests were excavated 3 to 5 days after the first emergence. green sea turtles nest on beaches in Taiwan (Chen and Cheng, 1995; On Wan-an Island, hatchlings were dissected on the day they were col- Cheng, 2011).