Relationship between tropical cyclones and the distribution of sea turtle nesting grounds

Journal of Biogeography (J. Biogeogr.) (2011) 38, 1886 1896 ORIGINAL ARTICLE Relationship between tropical cyclones and the distribution of sea turtle nesting grounds Mariana M. P. B. Fuentes 1 *, Brooke L. Bateman 2 and Mark Hamann 3 1 ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Qld 4811, Australia, 2 Centre for Tropical Biodiversity and Climate Change, School of Marine and Tropical Biology, James Cook University, Townsville, Qld 4811, Australia, 3 School of Earth and Environmental Sciences, James Cook University, Townsville, Qld 4811, Australia ABSTRACT Aim This study examines the relationship between the distribution of existing sea turtle nesting sites and historical patterns of tropical cyclone events to investigate whether cyclones influence the current distribution of sea turtle nesting sites. The results, together with information on predicted cyclone activity and other key environmental variables, will help in the identification and prediction of future nesting sites for sea turtles as changes to the coastal environment continue. Location Queensland, Australia. Methods We used data on the nesting distribution of seven populations of four species of sea turtles [green (Chelonia mydas), flatback (Natator depressus), hawksbill (Eretmochelys imbricata) and loggerhead (Caretta caretta)] from the eastern Queensland coast, and tropical cyclone track data from 1969 to 2007 to explore the relationship between (1) sea turtle nesting phenology and cyclone season, and (2) sea turtle nesting sites and cyclone distribution. Furthermore, using two green turtle populations as a case study, we investigated the relationship between cyclone disturbance and sea turtle reproductive output, nesting site and season. Bootstrapping was used to explore if current sea turtle nesting sites are located in areas with lower or higher cyclone frequency than areas where turtles are currently not nesting. Results All populations of sea turtles studied here were disturbed by cyclone activity during the study period. The exposure (frequency) of tropical cyclones that crossed each nesting site varied greatly among and within the various sea turtle populations. This was mainly a result of the spatial distribution of each population s nesting sites. Bootstrapping indicated that nesting sites generally have experienced lower cyclone activity than other areas that are available for nesting. *Correspondence: Mariana M. P. B. Fuentes, ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville, Qld, 4811, Australia. E-mail: mariana.fuentes@jcu.edu.au Main conclusions Tropical cyclones might have been sufficiently detrimental to sea turtle hatching success on the eastern Queensland coast that through a natural selection process turtles in this region are now nesting in areas with lower cyclone activity. Therefore, it is important that future studies that predict climate or range shifts for sea turtle nesting distributions consider future cyclone activity as one of the variables in their model. Keywords Australia, conservation management, disturbance, evolutionary selection, nesting sites, Queensland, sea turtles, species distribution, tropical cyclones. INTRODUCTION Understanding the relationships between species, their environment, and factors that determine their distribution and abundance has historically been a goal of ecological theory (MacArthur, 1972; Guisan & Zimmermann, 2000). The topic continues to be pertinent, as ongoing environmental change is likely to alter species ranges and shift their patterns of habitat 1886 http://wileyonlinelibrary.com/journal/jbi doi:10.1111/j.1365-2699.2011.02541.x

Sea turtle nesting sites and tropical cyclones use (Thomas et al., 2004; Malcolm et al., 2006). Thus there is a need to identify factors that have shaped existing species habitat interactions, future refugia and/or optimal areas for key species. This is particularly relevant for climate-related influences on the environment. Factors that determine the distribution of species are still not well understood, especially for marine species of conservation concern, such as sea turtles (Hamann et al., 2010; Santana-Garcon et al., 2010). Sea turtles lay their eggs in sandy coastal areas and on islands. The selection of a nesting site by sea turtles is a protracted process. Adult sea turtles have natal homing and often return to nest in the geographic region where they hatched (Bowen & Karl, 2007; Lohmann et al., 2008). Thus, beaches that produce the most surviving hatchlings might also have the highest numbers of adults returning to nest (Putman et al., 2010a). Consequently, poor hatchling success over several turtle generations may act as a selective process. Sea turtles also have the ability to colonize new nesting sites in the same geographic region in response to unsuitable conditions. For example, coastal development adjacent to nesting sites in Zakynthos, Greece, resulted in loggerhead turtles shifting their nesting sites to nearby, undeveloped, beaches (Schofield et al., 2010). Some existing nesting sites for sea turtles are likely to become unsuitable egg incubators in the future, owing to a combination of temperature increases, rising sea level, frequent inundation and increased coastal development (Fuentes et al., 2011). As an adaptive response to coastal change, sea turtles are predicted to shift their nesting sites or nesting phenology (Hays et al., 2001; Fuentes et al., 2011). Indeed, sea turtles have adapted to past climatic changes by redistributing their nesting sites and developing new migratory routes (Hamann et al., 2007). Thus, from a conservation management perspective it is important to understand the factors that underpin the distribution of sea turtle nesting sites (Kikukawa et al., 1999). Although there have been several attempts to determine how first-time breeders select an optimal nesting site and which factors make a nesting beach suitable, the beach characteristics underlying the distribution of sea turtle nesting sites at a population scale are poorly understood (Mortimer, 1990; Santana-Garcon et al., 2010). Most of the studies that have attempted to elucidate factors that determine nesting beach selection by female turtles have investigated the biophysical characteristics of specific nesting beaches (i.e. temperature, beach dimensions, slope, vegetation cover, sand softness) (e.g. Mortimer, 1990; Kikukawa et al., 1999; Chen et al., 2007; Fuentes et al., 2010), with a few studies investigating humanrelated beach characteristics (Kikukawa et al., 1999), broaderscale climatic processes (Santana-Garcon et al., 2010) and ocean currents (Putman et al., 2010a,b). Parameters considered important to sea turtle beach selection include: sand properties, distance from human settlement, presence of nearshore lagoon systems, access to the beach and exposure to wind, wave energy and ocean currents (Mortimer, 1990; Kikukawa et al., 1999; Putman et al., 2010a,b; Santana-Garcon et al., 2010; Hamann et al., 2011). However, little is known about how broader-scale aperiodic impacts from disturbance events, such as tropical cyclones (tropical storms, typhoons and hurricanes, as per Holland, 1993), have shaped the distribution of sea turtle nesting sites (Hamann et al., 2010). Determining the role of disturbance from extreme weather events in species distributions is important, as these events can exert substantial selective pressure and thus play a major role in structuring the current and future distributions of individuals and ecosystems (Parmesan et al., 2000; Jentsch et al., 2007; Jentsch & Beierkuhnlein, 2008; Thibault & Brown, 2008). Many of the world s sea turtle nesting sites are impacted by tropical cyclones, either during or outside the nesting season (Van Houtan & Bass, 2007; Fuentes & Abbs, 2010). Although cyclones can be drivers of important accretion events and can play a role in the propagation of coasts and potential nesting areas (Nott, 2006), their main influence on sea turtles tends to be associated with aperiodic disturbance to nesting beaches. Over time-scales of multiple turtle generations, cyclones can alter the biophysical dynamics of sea turtle nesting habitats, and over shorter time-scales, cyclones can decrease the annual or seasonal nesting success of females and increase the localized mortality of incubating eggs through beach erosion (Milton et al., 1994; Martin, 1996; Pike & Stiner, 2007; Van Houtan & Bass, 2007). Therefore it is speculated that the frequent, aperiodic occurrence of tropical cyclones along coastal sites has negatively impacted the nesting and hatching of sea turtles and thus may have helped to shape current sea turtle population boundaries and evolutionary processes. We therefore investigated how the distribution of existing sea turtle nesting sites in eastern Queensland relates to patterns of tropical cyclone events. To achieve this we used data on the nesting distribution of seven populations of four species of sea turtles from the eastern Queensland coast an area with globally significant populations of sea turtles and frequent tropical cyclones to explore the relationship between (1) sea turtle nesting phenology and cyclone season, and (2) sea turtle nesting sites and cyclone distribution. Furthermore, using two green turtle populations as a case study, we investigated the relationship between cyclone disturbance and sea turtle reproductive output, nesting site and season. We aim to elucidate whether disturbance events are important factors in shaping the distribution of sea turtle nesting sites. This, together with predicted information on cyclone activity and other key environmental variables, will help us to identify and predict future nesting sites for sea turtles as changes to the coastal environment progress. MATERIALS AND METHODS Study area We focused our study on sea turtle populations nesting along Australia s Queensland coast and the adjacent islands of the Great Barrier Reef World Heritage Area and Torres Strait (Fig. 1). This geographic area has the highest occurrence of tropical cyclones in the east Australian region and contains Journal of Biogeography 38, 1886 1896 1887

M. M. P. B. Fuentes et al. Figure 1 Eastern Queensland, Australia, showing the locations of the seven populations of four species of sea turtle [green (Chelonia mydas), flatback (Natator depressus), hawksbill (Eretmochelys imbricata) and loggerhead (Caretta caretta)] nesting in the region. GC, Gulf of Carpentaria; ngbr, northern Great Barrier Reef; CS, Coral Sea; sgbr, southern Great Barrier Reef; EA, Eastern Australia; SP, South Pacific. globally significant populations of green [Chelonia mydas (Linnaeus, 1758)], flatback [Natator depressus (Garman, 1880)], hawksbill [Eretmochelys imbricata (Linnaeus, 1766)] and loggerhead [Caretta caretta (Linnaeus, 1758)] turtles. Seven genetically distinct sea turtle populations nest within this area, namely the northern Great Barrier Reef (ngbr), the southern Great Barrier Reef (sgbr) and the Coral Sea (CS) green turtle populations; the Gulf of Carpentaria (GC) and the eastern Australian (EA) flatback populations; the hawksbill population and the South Pacific loggerhead population (SP) (Limpus, 2007) (Fig. 1). Sea turtle nesting data Information on the distribution and phenology of nesting for each turtle population was gathered from published material (Limpus, 1971; Limpus et al., 1993, 2003; Dobbs et al., 1999; Limpus & Miller, 2000; Harvey et al., 2005) and used to generate a GIS layer of nesting locations. For the purpose of this study, rookeries with < 10 nesting turtles per year are considered trivial and are not represented in the nesting layers. The nesting turtle datasets we used represent more than 100 years of similar patterns in nesting site distributions for the species, which represents at least three turtle generations (Moorhouse, 1933). There is evidence that some of the nesting sites, such as Raine Island, have been used for more than 1500 years (Limpus et al., 2003). Historical frequency of tropical cyclones The track of each tropical cyclone that crossed the eastern Queensland coast and adjacent islands from 1969/70 to 2006/07 was obtained from the Australian Bureau of Meteorology (BOM: http://www.bom.gov.au/weather/cyclone/ 1888 Journal of Biogeography 38, 1886 1896

Sea turtle nesting sites and tropical cyclones tc-history.shtml). Data prior to the 1968/69 season were not used because the lack of direct observations in this earlier period resulted in positional inaccuracies of up to 250 km (Holland, 1981). However, the dataset used probably represents the spatial patterns of cyclones for the past thousands of years. Research using radiocarbon dating indicates that cyclone frequency has been statistically constant over the last 5000 years in north Queensland (see Hayne & Chappell, 2001). To determine the cumulative frequency of tropical cyclones within the study region the following steps were undertaken: (1) each vector cyclone track from 1969/70 to 2006/07 was buffered by 40 km (which represents the average cyclone eye and the area that would be most affected by severe winds) (Australian Bureau of Meteorology, 2008), (2) the vector cyclone layers were then converted into a raster layer with a resolution of 0.25, (3) each raster layer (representing a single cyclone path) was reclassified to 1 for the buffered cyclone track and 0 for the remaining area, and (4) all the raster layers were added using the Raster Calculator in ArcGIS 9.1 (ESRI, Redlands, CA, USA). This resulted in a raster layer with 0.25 grid cells with values of the frequency of tropical cyclones for the study period. We determined the frequency of tropical cyclones that crossed each sea turtle nesting site using zonal statistics and compared the frequency of tropical cyclones between the various nesting sites using a one-way ANOVA (spss 18). Bootstrapping was used to explore whether current turtle nesting sites are located in areas with lower or higher cyclone frequency than areas where turtles are currently not nesting. For this, a subset of 129 random points (approximate number of nesting sites used by the populations being studied) were projected 100 times in areas where sea turtles are not currently nesting but have the physical characteristics that would allow nesting (i.e. sandy beaches). Values for the frequency of tropical cyclones that passed each set of control points during the study period (38 seasons) were averaged and compared with the average frequency of cyclones per nesting ground during the study period. All spatial analyses were performed using ArcGIS 9.1. Disturbance by cyclones We explore how the number of sea turtle eggs exposed to cyclones varies as a function of (1) the number of females breeding in a particular season, (2) the timing of the cyclone hits to a nesting site, and (3) the importance of the nesting site, using the ngbr and sgbr green turtle populations as a case study. For the ngbr population, we assumed that 131,000, 41,000 and 22,900 turtles nest during a high, average and low nesting season, respectively, and that each turtle lays on average 600 eggs during a nesting season (six clutches of 100 eggs) (as per Limpus et al., 2003). The proportion of clutches laid each month (October to March) and the average incubation period (56 days) was calculated from the 1979 80 nesting dataset from Bramble Cay (Limpus et al., 2001). The season was an average season in terms of numbers of turtles breeding (Limpus et al., 2003). The ngbr green turtle population nests at several sites, with some having more importance (proportional to the number of turtles nesting) than others, so we gave weights to the key nesting sites (n = 7) (as per Fuentes et al., 2011). Weights are based on the percentage of nesting that occurs at each site in relation to the overall nesting across the seven sites, which cumulatively represent 99% of the nesting for this population. A similar approach was used with the sgbr population, except that we used nesting abundance data from one nesting ground (Heron Island) rather than for the whole population as we did with the ngbr population. This was necessary because population-level estimates for the sgbr green turtles are not available for individual years. However, the nesting distribution of green turtles in the sgbr population is more evenly spread among nesting grounds, and thus there are unlikely to be differences in phenology or biological traits between nesting grounds for this population (Limpus et al., 1984). We assumed that 50, 540 and 1030 turtles nest at Heron Island during a low, average and high nesting season (Limpus et al., 2003), respectively, that each turtle lays five clutches of 115 eggs during a nesting season (Hamann, 2002), and that eggs take an average of 64 days to incubate (Limpus, 2007). The proportion of clutches laid each month (October to March) was calculated for a slightly lower than average nesting season (1997 98, n = 289 females, which is < 1 SD from the mean) because this is the only season for which monitoring data are available for the entire season (Hamann, 2002). RESULTS Sea turtle nesting activity The distribution of nesting sites varied with species (Fig. 1). No substantial nesting for any species occurs from Proserpine (20 S) north to Cooktown (15 S) (Fig. 1). In terms of phenology, year-round nesting, with a distinct seasonal peak, occurs for the GC flatback, ngbr green and the hawksbills. All the other populations have well-defined nesting seasons and, with slight species variations, they nest between October and April (Limpus, 1971; Limpus & Reed, 1985; Limpus & Miller, 2000; Limpus et al., 2003; Fuentes & Abbs, 2010). Cyclone activity A total of 172 tropical cyclones passed through the study area between 1969/70 and 2006/07 (38 cyclone seasons), with an average of 4.52 (± 1.78; range 2 9) tropical cyclones each year. The frequency of cyclones that passed through the study region decreased with time (regression, P = 0.054, r = )0.3236, F = 3.978, d.f. = 1). Cyclone occurrence was higher from Proserpine (20 S) northwards to Cooktown (15 S), and into the Gulf of Carpentaria, with concentrated activity on Cape York Peninsula and in the Coral Sea (Fig. 2). Tropical cyclones Journal of Biogeography 38, 1886 1896 1889

M. M. P. B. Fuentes et al. Figure 2 Eastern Queensland, Australia, showing the locations of the seven populations of four species of sea turtle [green (Chelonia mydas), flatback (Natator depressus), hawksbill (Eretmochelys imbricata) and loggerhead (Caretta caretta)] nesting in the region and the sum frequency of cyclone activity from 1969/70 to 2006/07. GC, Gulf of Carpentaria; ngbr, northern Great Barrier Reef; CS, Coral Sea; sgbr, southern Great Barrier Reef; EA, Eastern Australia; SP, South Pacific. always occurred between November and May, with peak cyclone activity occurring during January and February (Fig. 3). Historical exposure of sea turtle nesting sites to tropical cyclones Exposure to tropical cyclones, in terms of frequency of hits, varied among and within the various sea turtle populations (ANOVA, P < 0.00, d.f. = 6, F = 4.401) (Fig. 2). Based on cyclone data from 1969 70 and 2006 07, nesting sites used by the CS green turtle population have been the most disturbed by tropical cyclones (post hoc Tukey s honestly significant difference (HSD) test, P = 0.01 with all groups) (average of 0.22 ± 0.l hits a year per nesting site). This is high when compared with the populations of sea turtles in the Torres Strait (0.044 to 0.060 cyclone hits a year per nesting site) and the southern Great Barrier Reef regions (0.090 to 0.098 cyclone hits a year per nesting site) (Table 1). Peak nesting for the three green turtle populations (December January) and the hawksbill population (January February) coincides with peak cyclone activity (January February). While peak nesting for the SP loggerhead and EA flatback populations occurs in late and early December, respectively, which is prior to the peak cyclone season, a high proportion of their eggs are still incubating during the peak of the cyclone season (Fuentes & Abbs, 2010). In contrast, the peak of the nesting season for the GC flatbacks occurs outside the cyclone season (July September) (see Fig. 3). Bootstrapping indicated that sea turtles nesting in eastern Queensland are generally nesting in sites with lower cyclone activity than other areas that are available for nesting. The mean cyclone frequency at nesting sites (3.59 hits) was 1890 Journal of Biogeography 38, 1886 1896

Sea turtle nesting sites and tropical cyclones Figure 3 Eastern Queensland, Australia, showing the distribution of nesting sites of four species of sea turtle [green (Chelonia mydas), flatback (Natator depressus), hawksbill (Eretmochelys imbricata) and loggerhead (Caretta caretta)] and monthly cyclone frequency from 1969/70 to 2006/07. The turtle nesting sites are in the same locations as in Figs 1 & 2. Journal of Biogeography 38, 1886 1896 1891

M. M. P. B. Fuentes et al. Table 1 Average cyclone frequency at nesting sites used by each of the seven sea turtle populations that nest along the eastern Queensland coast, Australia. Sea turtle population Average annual frequency of cyclones that passed through nesting sites a year ± SE Average number of tropical cyclones that crossed each nesting site from 1969/70 to 2006/07 Region Northern Great Barrier Reef (GBR) 0.060 ± 0.06 2.28 Torres Strait green turtle (Chelonia mydas) Hawksbill turtle (Eretmochelys imbricata) 0.044 ± 0.04 1.67 Torres Strait Gulf of Carpentaria flatback turtle 0.055 ± 0.10 2.09 Torres Strait (Natator depressus) Southern GBR green turtle 0.098 ± 0.05 3.72 Southern Great Barrier Reef South Pacific loggerhead turtle 0.096 ± 0.04 3.64 Southern Great Barrier Reef (Caretta caretta) Eastern Australian flatback turtle 0.090 ± 0.05 3.42 Southern Great Barrier Reef Coral Sea green turtle 0.220 ± 0.10 8.36 Coral Sea significantly lower (one-way ANOVA, P < 0.00, d.f. = 100, F = 10.545, post hoc Tukey s HSD test, P < 0.00 with all groups) than the mean frequency for the random subset of points generated (which ranged from 4.91 to 5.96 hits). Variation in cyclone disturbance Results from our conceptual model, which used two green turtle populations as a case study, indicated that the magnitude of disturbance by cyclones varies greatly in accordance with (1) the number of females breeding in a particular season, (2) the time of year at which the cyclone hits a nesting ground, and (3) for the ngbr population, the importance of the nesting site that is hit by a cyclone (Fig. 4a,b). More specifically, the exposure of egg clutches to cyclones can be 20.6 (sgbr) and 5.7 (ngbr) times higher during a high nesting season than during a low nesting season (Fig. 4a,b). Similarly, if a cyclone hits a green turtle nesting site when the majority of eggs are incubating (January), it will expose 5 (sgbr) and 3 (ngbr) times more eggs to cyclones than if a cyclone hit a nesting site at the beginning of the nesting season in November (Fig. 4a,b). For the ngbr green turtle population, the exposure of eggs to cyclones can be 100 times higher if it hits the main nesting ground (Raine Island) rather than a trivial nesting ground such as Milman Island (Fig. 4a). DISCUSSION All populations of sea turtles that breed in eastern Queensland were disturbed by cyclone activity between 1969 and 2007. The frequency of tropical cyclone disturbance to sites varied greatly among and within the various sea turtle populations. The nesting sites for the Coral Sea green turtle population were the most disturbed (in terms of frequency of hits) by tropical cyclones, while the nesting sites for the hawksbill population were the least disturbed by cyclone activity. Variation in exposure to cyclone activity was mainly a result of the spatial distribution of each population s nesting sites. Within the study period, eastern Queensland s nesting sites were hit by a cyclone, on average, every 4 years in the Coral Sea, every 9 years in the Great Barrier Reef and only once in Torres Strait. Disturbance by tropical cyclones occurred at the time when most of the eggs were incubating (January February), which is when the greatest impact to sea turtles can occur. In a high nesting year, the effects of disturbance to turtle eggs can be about 20.6 (sgbr) and 5.7 (ngbr) times larger during the peak of incubation (January to February) than during the beginning of the nesting season (October and November). At a key nesting site, such as Raine Island, this could equate to a difference of more than 116 million eggs being exposed to cyclones. Other factors that may influence the impact of tropical cyclones on the stability of sea turtle populations include: (1) the location and biophysical characteristics of the nesting site (i.e. cays/mainland, vegetation, sedimentology and morphology) (Fuentes & Abbs, 2010), (2) tidal state, level of storm surge and wave height during cyclone activity (Boswood & Mohoupt, 2007), (3) the direction of the wind, and (4) the characteristics of the cyclones, such as central pressure, speed, diameter and intensity (Flood, 1986). Ultimately, the overall impact of a particular cyclone on a population of sea turtles will be strongly linked to the importance of the nesting site (in terms of the percentage of turtles from a population that nests on that site) that the cyclone hits. This was clearly demonstrated by our conceptual model, which showed that if a cyclone hits Raine Island and Moulter Cay (separated by c. 20 km), which are the main nesting sites for the ngbr green turtle population (Limpus et al., 2003), the impact could be of the order of 100 times larger than if it hit a trivial nesting site such as Milman Island. If a cyclone were to hit during a high nesting season and at the point of peak of egg incubation, this could mean a potential loss of 118 million eggs. This disparity in the magnitude of disturbance from a cyclone hitting different nesting sites for a particular population is significant only 1892 Journal of Biogeography 38, 1886 1896

Sea turtle nesting sites and tropical cyclones (a) (b) Figure 4 Number of green turtle (Chelonia mydas) eggs exposed to cyclone disturbance at (a) key nesting sites used by the northern Great Barrier Reef (ngbr) green turtle population in Torres Strait and northern Great Barrier Reef, Australia, and (b) Heron Island, a key nesting site for the southern Great Barrier Reef (sgbr) green turtle population during various nesting seasons and months. for the ngbr green and the GC flatback turtles. This is because most of the breeding for the other populations, studied here, is more concentrated (i.e. less latitudinal spread of sites), and the nesting abundance more evenly spread among individual sites, and hence cyclones are more likely to impact each of the nesting grounds used by these other populations similarly. For green turtles, cyclone disturbance will also vary depending on how many females are breeding in a particular nesting season. The number of green turtles nesting each year fluctuates seasonally, and the number of females breeding correlates with the Southern Oscillation index 18 months before the breeding season (Limpus & Nicholls, 2000; Limpus et al., 2003). Consequently, the impact of a few cyclones in a high nesting year may be greater than that of more cyclones in a low nesting year. The most destructive years are those with a high frequency of tropical cyclones that coincide with a high nesting density/abundance (Pike & Stiner, 2007). In Queensland this was the case in 1984/85, 1989/90, 1995/96 and 1996/ 97, when green turtle nesting densities (Limpus et al., 2003) and cyclone frequencies (Australian Bureau of Meteorology, 2008) were both high. Indeed, during 1995/96, the Coral Sea green turtle population had one of its highest nesting years (approximately 2300 clutches), and this was also when three tropical cyclones crossed the Coral Sea, all in February and March when the majority of the eggs were incubating (Harvey et al., 2005). Tropical cyclones are a disturbance to sea turtle populations as they are discrete events in time that increase the localized mortality of eggs, change resources and nesting area availability, and disrupt population structure (Fuentes & Abbs, 2010). Despite sea turtle populations having life histories that are capable of withstanding the loss of a single cohort of eggs/hatchlings (Heppell et al., 1996), frequent disturbances to hatching (and thus low hatching success over time) owing to tropical cyclones can have a profound influence on the longer-term reproductive output of a population, potentially exerting substantial selective pressures (Levin & Paine, 1974; Wootton, 1998; Zimmermann et al., 2009). It seems plausible that this was the case in our study region. Our study found that sea turtles along the eastern Queensland coast nest in areas with a lower historical incidence of tropical cyclones. It is possible that areas with a high incidence of tropical cyclones might have substantially affected nesting and/or hatching success over time, reducing the number of turtles that return to nest in these areas over several generations. Thus, cyclone activity may have been an important factor in shaping the distribution of sea turtle nesting sites along the eastern Queensland coast. Consequently, it is important that future studies that predict Journal of Biogeography 38, 1886 1896 1893

M. M. P. B. Fuentes et al. climate- or development-induced range shifts for sea turtle nesting distributions consider current and future cyclone activity as one of the variables in their model. However, studies on species distributions and extreme weather events have focused mainly on terrestrial environments and have only incorporated heat waves and droughts in spatial predictions (Brooke Bateman, Jeremy VanDerWal & Chris Johnson, unpublished data), with no study to date incorporating historical and predicted information of cyclone activity (Franklin, 2010; Luja & Rodríguez-Estrella, 2010). It is likely that climate extremes such as tropical cyclones play a greater role in influencing species and their distributions than gradual changes in long-term climate averages (Zimmermann et al., 2009), particularly within physiologically stressful conditions. As cyclone frequency, intensity, distribution and seasonality are predicted to alter with climate change (Walsh & Ryan, 2000; Webster et al., 2005; Abbs et al., 2007; Leslie et al., 2007; Kuleshov et al., 2008), information on how such events affect species and their distributions, and on the ability of organisms to recover from them, is crucial in planning for future changes in climate and coastal change (Dale et al., 1998). Tropical cyclones are seldom included in management plans because of their aperiodic and unpredictable nature. Nevertheless, our data indicate they should be considered and can be incorporated into species and ecosystem management plans (Dale et al., 1998; Sutherland et al., 2009). Managers can choose to influence (1) the system prior to the disturbance, (2) the system after the disturbance, or (3) the recovery process. Prior to the disturbance, managers can alter the vulnerability of the system, or change how it will respond to the disturbance. After the disturbance, managers can manage the system or aid the ongoing process of recovery (Dale et al., 1998). Regarding sea turtles and tropical cyclones along the eastern Queensland coast, managers can relocate eggs to areas that are safer prior to cyclone activity (as occurs for the main nesting site for the SP loggerhead), and, postcyclone, they can recover exposed clutches of eggs that still have a chance of surviving, rescue disoriented hatchlings and stranded sea turtles, and manage the nesting site so that it returns to pre-disturbance conditions (e.g. manage erosion, flooding). Hence, tropical cyclones and their historical and future influence on current distributions of sea turtles should be investigated in other regions that are important for sea turtles in order to provide key information for efficient longer-term management. ACKNOWLEDGEMENTS Special thanks go to J. Santana-Garcon for assistance in collecting information on the nesting seasonality and location for each population. M.M.P.B.F. was funded by the Australian Research Council Super Science Fellowship. M.H. was supported by the Australian Government s Marine and Tropical Sciences Research Facility. We are extremely grateful to the comments provided by C.J. Limpus and J. Moloney during the initial drafts of this manuscript. REFERENCES Abbs, D.J., Timbal, B., Rafter, A.S. & Walsh, K.J.E. (2007) Severe weather. Climate change in Australia (ed. by K.B. Pearce, P.N. Holper, M. Hopkins, W.J. Bouma, P.H. Whetton, K.J. Hennessy and S.B. Power), pp. 102 106. Technical report 2007, CSIRO Marine and Atmospheric Research, Aspendale, Australia. Australian Bureau of Meteorology (2008) Tropical cyclone database for Australia. Available at: http://www.bom.gov.au/ index.shtml (accessed in February 2008). Boswood, P.K. & Mohoupt, J. (2007) Tropical Cyclone Larry post cyclone, coastal field investigation. Environmental Protection Agency, Brisbane. Bowen, B.W. & Karl, S.A. (2007) Population genetics and phylogeography of sea turtles. Molecular Ecology, 16, 4886 4907. Chen, H.C., Cheng, I.J. & Hong, E. (2007) The influence of beach environment on the digging success and nest site distribution of the green turtle, Chelonia mydas, on Wan-an Island, Penghu Archipelago, Taiwan. Journal of Coastal Research, 23, 1277 1286. Dale, V.H., Lugo, A.E., MacMahon, J.A. & Pickett, S.T.A. (1998) Ecosystem management in the context of large, infrequent disturbances. Ecosystems, 1, 546 557. Dobbs, K.A., Miller, J.D., Limpus, C.J. & Landry, A.M.J. (1999) Hawksbill turtle, Eretmochelys imbricata, nesting at Milman Island, northern Great Barrier Reef, Australia. Chelonian Conservation Biology, 3, 344 361. Flood, P.G. (1986) Sensitivity of coral cays to climatic variations, southern Great Barrier Reef, Australia. Coral Reefs, 5, 13 18. Franklin, J. (2010) Moving beyond static species distribution models in support of conservation biogeography. Diversity and Distributions, 16, 321 330. Fuentes, M. & Abbs, D. (2010) Effects of projected changes in tropical cyclone frequency on sea turtles. Marine Ecology Progress Series, 412, 283 292. Fuentes, M.M.P.B., Dawson, J.L., Smithers, S.G., Hamann, M. & Limpus, C.J. (2010) Sedimentological characteristics of key sea turtle rookeries: potential implications under projected climate change. Marine and Freshwater Research, 61, 464 473. Fuentes, M.M.P.B., Limpus, C.J. & Hamann, M. (2011) Vulnerability of sea turtle nesting grounds to climate change. Global Change Biology, 17, 140 153. Guisan, A. & Zimmermann, N.E. (2000) Predictive habitat distribution models in ecology. Ecological Modelling, 135, 147 186. Hamann, M. (2002) Reproductive cycles, interrenal gland function and lipid mobilisation in the green sea turtle (Chelonia mydas). PhD Thesis, University of Queensland, Brisbane. Hamann, M., Limpus, C.J. & Read, M.A. (2007) Vulnerability of marine reptiles in the Great Barrier Reef to climate change. Climate change and the Great Barrier Reef: a 1894 Journal of Biogeography 38, 1886 1896

Sea turtle nesting sites and tropical cyclones vulnerability assessment (ed. by J.E. Johnson and P.A. Marshall), pp. 465 496. Great Barrier Reef Marine Park Authority and Australia Greenhouse Office, Hobart, Tasmania. Hamann, M., Godfrey, M., Seminoff, J. et al. (2010) Global research priorities for sea turtles: informing management and conservation in the 21st century. Endangered Species Research, 11, 245 269. Hamann, M., Grech, A., Wolanski, E. & Lambrechts, J. (2011) Modelling the fate of marine turtle hatchlings. Ecological Modelling, 222, 1515 1521. Harvey, T., Townsend, S., Kenyon, N. & Redfern, G. (2005) Monitoring of nesting sea turtles in the Coringa-Herald National Nature Reserve (1991/92 to 2003/2004 nesting seasons). Report to the Australian Department of Environment and Heritage, Canberra. Hayne, M. & Chappell, J. (2001) Cyclone frequency during the last 5000 years at Curacoa Island, northern Queensland, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology, 168, 207 219. Hays, G.C., Ashworth, J.S., Barnsley, M.J., Broderick, A.C., Emery, D.R., Godley, B.J., Henwood, A. & Jones, E.L. (2001) The importance of sand albedo for the thermal conditions on sea turtle nesting beaches. Oikos, 93, 87 94. Heppell, S.S., Limpus, C.J., Crouse, D.T., Frazer, N.B. & Crowder, L.B. (1996) Population model analysis for the loggerhead sea turtles, Caretta caretta, in Queensland. Wildlife Research, 23, 143 159. Holland, G.J. (1981) On the quality of the Australian tropical cyclone data base. Australian Meteorological Magazine, 29, 169 181. Holland, G.J. (1993) Ready Reckoner. Global guide to tropical cyclone forecasting (ed. by G.J. Holland), pp. 19 38. World Meteorological Organization, Geneva. Jentsch, A. & Beierkuhnlein, C. (2008) Research frontiers in climate change: effects of extreme meteorological events on ecosystems. Comptes Rendus Geosciences, 340, 621 628. Jentsch, A., Kreyling, J. & Beierkuhnlein, C. (2007) A new generation of climate-change experiments: events, not trends. Frontiers in Ecology and the Environment, 5, 365 374. Kikukawa, A., Kamezaki, N. & Ota, H. (1999) Factors affecting nesting beach selection by loggerhead turtles (Caretta caretta): a multiple regression approach. Journal of Zoology, 249, 447 454. Kuleshov, Y., Qi, L., Fawcett, R. & Jones, D. (2008) On tropical cyclone activity in the Southern Hemisphere: trends and the ENSO connection. Geophysical Research Letters, 35, L14S08. doi:10.1029/2007gl032983. Leslie, L.M., Karoly, D.J., Leplastrier, M. & Buckley, B.W. (2007) Variability of tropical cyclones over the southwest Pacific Ocean using a high resolution climate model. Meteorology and Atmospheric Physics, 97, 171 180. Levin, S.A. & Paine, R.T. (1974) Disturbance, patch formation, and community structure. Proceedings of the National Academy of Sciences USA, 71, 2744 2747. Limpus, C.J. (1971) The flatback turtle, Chelonia depressa, Garman, in southeast Queensland, Australia. Herpetologica, 27, 431 446. Limpus, C.J. (2007) A biological review of Australian marine turtle species. Queensland Environmental Protection Agency, Brisbane. Limpus, C.J. & Miller, J.D. (2000) Final report for Australian hawksbill turtle population dynamics project. Japan Bekko Association to Queensland Parks and Wildlife Service, Brisbane. Limpus, C.J. & Nicholls, N. (2000) ENSO regulation of Indo- Pacific green turtle populations. Applications of seasonal climate forecasting in agricultural and natural ecosystems the Australian experience (ed. by G.L. Hammer, N. Nicholls and C. Mitchell), pp. 399 408. Kluwer Academic Publishers, Dordrecht. Limpus, C.J. & Reed, P.C. (1985) The loggerhead turtle, Caretta caretta, in Queensland (Australia): observations on interesting behavior. Australian Wildlife Research, 12, 535 540. Limpus, C.J., Fleay, A. & Guinea, M. (1984) Sea turtles of the Capricorn section, Great Barrier Reef. The Capricorn section of the Great Barrier Reef: past, present and future (ed. by W.T. Ward and P. Saenger), pp. 61 78. The Royal Society of Queensland, Brisbane. Limpus, C.J., Couper, P.J. & Couper, K.L.D. (1993) Crab Island revisited: reassessment of the world s largest flatback turtle rookery after twelve years. Memoirs of the Queensland Museum, 33, 277 289. Limpus, C.J., Carter, D. & Hamann, M. (2001) The green turtle, Chelonia mydas, in Queensland, Australia: the Bramble Cay rookery in the 1979 1980 breeding season. Chelonian Conservation and Biology, 4, 34 46. Limpus, C.J., Miller, J., Parmenter, J. & Limpus, D.J. (2003) The green turtle, Chelonia mydas, population of Raine Island and the northern Great Barrier Reef 1843 2001. Memoirs of the Queensland Museum, 49, 349 440. Lohmann, K.J., Putman, N.F. & Lohmann, C.M.F. (2008) Geomagnetic imprinting: a unifying hypothesis of natal homing in salmon and sea turtles. Proceedings of the National Academy of Sciences USA, 105, 19096 19101. Luja, V.H. & Rodríguez-Estrella, R. (2010) Are tropical cyclones sources of natural selection? Observations on the abundance and behavior of frogs affected by extreme climatic events in the Baja California Peninsula, Mexico. Journal of Arid Environments, 74, 1345 1347. MacArthur, R.H. (1972) Geographical ecology. Harper & Row, New York. Malcolm, J.R., Liu, C.R., Neilson, R.P., Hansen, L. & Hannah, L. (2006) Global warming and extinctions of endemic species from biodiversity hotspots. Conservation Biology, 20, 538 548. Martin, R.E. (1996) Storm impacts on loggerhead turtle reproductive success. Marine Turtle Newsletter, 73, 10 12. Milton, S.L., Leone-Kabler, S., Schulman, A.A. & Lutz, P.L. (1994) Effects of Hurricane Andrew on the sea turtle nesting Journal of Biogeography 38, 1886 1896 1895

M. M. P. B. Fuentes et al. beaches of South Florida. Bulletin of Marine Science, 54, 974 981. Moorhouse, F.W. (1933) Notes on the green turtle (Chelonia mydas). Great Barrier Reef Committee Report, 4, 1 22. Mortimer, J.A. (1990) The influence of beach sand characteristics on the nesting behavior and clutch survival of green turtles (Chelonia mydas). Copeia, 1990, 802 817. Nott, J. (2006) Tropical cyclones and the evolution of the sedimentary coast of Northern Australia. Journal of Coastal Research, 22, 49 62. Parmesan, C., Root, T.L. & Willig, M.R. (2000) Impacts of extreme weather and climate on terrestrial biota. Bulletin of the American Meteorological Society, 81, 443 450. Pike, D.A. & Stiner, J.C. (2007) Sea turtle species vary in their susceptibility to tropical cyclones. Oecologia, 153, 471 478. Putman, N.F., Bane, J.M. & Lohamann, K.J. (2010a) Sea turtle nesting distributions and oceanographic constraints on hatchling migration. Proceedings of the Royal Society B: Biological Sciences 277, 3631 3637. Putman, N.F., Shay, T.J. & Lohamann, K.J. (2010b) Is the geographic distribution of nesting in the Kemp s Ridley turtle shaped by the migratory needs of offspring? Integrative and Comparative Biology, 50, 305 314. Santana-Garcon, J.S., Grech, A., Moloney, J. & Hamann, M. (2010) Relative Exposure Index: an important factor in sea turtle nesting distribution. Aquatic Conservation: Marine and Freshwater Ecosystems, 20, 140 149. Schofield, D.G., Hobson, V.J., Lilley, M.K.S., Katselidis, K.A., Bishop, C.M., Brown, P. & Hays, G.C. (2010) Inter-annual variability in the home range of breeding turtles: implications for current and future conservation management. Biological Conservation, 143, 722 730. Sutherland, W.J., Adams, W.M., Aronson, R.B. et al. (2009) One hundred questions of importance to the conservation of global biological diversity. Conservation Biology, 23, 557 567. Thibault, K.M. & Brown, J.H. (2008) Impact of an extreme climatic event on community assembly. Proceedings of the National Academy of Sciences USA, 105, 3410 3415. Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C., Erasmus, B.F.N., de Siqueira, M.F., Grainger, A., Hannah, L., Hughes, L., Huntley, B., van Jaarsveld, A.S., Midgley, G.F., Miles, L., Ortega-Huerta, M.A., Peterson, A.T., Phillips, O.L. & Williams, S.E. (2004) Extinction risk from climate change. Nature, 427, 145 148. Van Houtan, K.S. & Bass, O.L. (2007) Stormy oceans are associated with declines in sea turtle hatching. Current Biology, 17, R590 R591. Walsh, K.J.E. & Ryan, B.F. (2000) Tropical cyclone intensity increase near Australia as a result of climate change. Journal of Climate, 13, 3029 3036. Webster, P.J., Holland, G.J., Curry, J.A. & Chang, H.R. (2005) Changes in tropical cyclone number, duration, and intensity in a warming environment. Science, 309, 1844 1846. Wootton, J.T. (1998) Effects of disturbance on species diversity: a multitrophic perspective. The American Naturalist, 152, 803 825. Zimmermann, N.E., Yoccoz, N.G., Edwards, T.C., Meier, E.S., Thuiller, W., Guisan, A., Schmatz, D.R. & Pearman, P.B. (2009) Climatic extremes improve predictions of spatial patterns of tree species. Proceedings of the National Academy of Sciences USA, 106, 19723 19728. BIOSKETCHES Mariana Fuentes is a postdoctoral fellow at the ARC Centre of Excellence for Coral Reef Studies at James Cook University. Her broad scientific interest lies in providing information to aid the conservation and management of threatened marine mega-fauna in a changing climate. Brooke Bateman is a research fellow at the Centre for Tropical Biodiversity and Climate Change at James Cook University. Her current research interests are in spatial ecology and conservation, with particular focus on threatened species and marsupials. Mark Hamann is a senior lecturer in environmental science at James Cook University. His research activities are primarily in the field of applied ecology in relation to improving the management of marine wildlife. Editor: Craig McClain 1896 Journal of Biogeography 38, 1886 1896