For thousands of years, marine turtles have exhibited high plasticity in adapting to environmental variations, making them globally resilient species (Fuentes et al. 2013). However, the International Union for Conservation of Nature (IUCN; http://www.iucnredlist.org/) classifies the hawksbill turtle (Eretmochelys imbricata) as Critically Endangered (Mortimer and Donnelly 2008). The hawksbill is also considered as an indicator species because of its sensitivity and vulnerability to changes in the habitats it occupies. Hawksbill populations' adaptations to past and present environmental changes make this species an interesting case study for ecological and habitat restoration assessments. The hawksbill is also a Priority Species for Conservation in Mexico, making it the focus of many conservation efforts (Comisión Nacional de Áreas Naturales Protegidas 2009).

Nest selection on nesting beaches is among the most important aspects of hawksbill reproductive biology (Witt et al. 2010; Huerta-Rodríguez et al. 2013; Kelly et al. 2017). Temperature is one of several environmental parameters reported as determining reproductive success in this species; it affects embryo survival and regulates the relative number of male and female offspring produced (Kamel and Mrosovsky 2006a).

Other nesting beach physical parameters that influence sea turtle reproductive success include vegetation community type and structure on the beach and in adjacent ecosystems, sand organic content, and sand sedimentological characteristics (Moreno-Casasola and Travieso-Bello 2006; Bolongaro-Crevenna-Recaséns et al. 2010; Ditmer and Stapleton 2012). Vegetation is particularly important because it shades nests, consequently affecting nest temperature (Mrosovsky et al. 1992; Kolbe and Jansen 2002; Kamel 2013). Beach extent and slope can also impact reproductive success because both parameters are associated with the capacity of female turtles to locate adequate nesting sites (Fish et al. 2005; Cuevas et al. 2010). These nesting habitat physical and environmental characteristics influence sea turtle reproductive success and therefore the degree to which populations are sustained (Kamel and Mrosovsky 2006a; Pfaller et al. 2008).

The western coast of the Yucatan Peninsula, Mexico, has extensive sandy beaches and is a critical hawksbill turtle nesting and hatching habitat. However, it is expected to experience strong impacts from the sea level rise predicted for coming decades (Grupo Intergubernamental de Expertos sobre el Cambio Climático [IPCC] 2013) and from increasing exposure to pressure from human activities (Ortíz-Pérez and Méndez-Linares 1999). As the main nesting area for hawksbill turtles in the Western Atlantic (Mortimer and Donnelly 2008), this region is considered strategic for marine turtle conservation by national governmental and international agencies (National Fish and Wildlife Foundation 2010; Campbell 2014; Guzmán and García 2015). Mexico has signed the Inter-American Convention for the Protection and Conservation of Sea Turtles (IAC 1996) and the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES 1973). Both conventions constitute commitments to contribute to strengthening conservation and protection actions throughout the country.

Chenkan beach in the state of Campeche is a Ramsar site (Convention on Wetlands [Ramsar Convention] 1971) which supports the third-largest hawksbill nesting population in the state (> 350 hawksbill nests per yr) (Huerta-Rodríguez and Muñoz 2009; Guzmán 2017). Despite the efforts of numerous nongovernmental organizations and government authorities to protect marine turtles, the habitats hawksbills use, such as Chenkan beach, are being modified by several anthropogenic threats including fishing and urban infrastructure development. This leads to habitat fragmentation and loss of vegetation cover on the sandy dunes backing nesting beaches (Comisión Nacional de Áreas Naturales Protegidas 2009). Dune vegetation cover is a critical component of beach ecosystems which contributes to stabilizing the sediment sea turtles use to build their nests and to regulating a nesting area's thermal regime (Nordstrom et al. 2000; Martínez et al. 2004; Hannan et al. 2007).

For several decades, most of the conservation efforts aimed at protecting marine turtles and their habitat have focused on the sandy beaches where they nest. The goal is to prevent population decline by guaranteeing offspring viability through in situ and ex situ beach management strategies and the release of emerged hatchlings to the sea (Comisión Nacional de Áreas Naturales Protegidas 2009). However, the aforementioned ongoing and predicted threats to nesting beaches along the Yucatan Peninsula threaten the long-term viability of the local hawksbill turtle nesting population.

The present study's objective was to assess the influence of nesting beach physical and environmental factors on reproductive parameters including nesting success, hatching success, and hatchling emergence success. Because preservation or restoration of natural habitat conditions on nesting beaches is expected to positively influence marine turtle reproductive success, the importance of protecting the natural beach habitat (encompassing the beach, dunes, and associated vegetation) is discussed in terms of how it may help decision-makers and authorities to design optimum conservation and restoration programs for this endangered species.

METHODS

Study Area

This study was conducted on the nesting beach of Chenkan, located on the western coast of the Yucatan Peninsula in Campeche state in the southeastern Gulf of Mexico. This beach is an important marine turtle nesting ground and has been a Ramsar site since 2003 (Huerta-Rodríguez and Munoz 2009). The beach is also adjacent the Laguna de Términos Flora and Fauna Protection Area (Fig. 1). Chenkan Beach is a highly dynamic area subject to beach erosion. The dominant beach profile consists of a pronounced slope and reduced beach width, with sand grain coarseness ranging from thick (2–5 mm) to intermediate (0.08–1.99 mm) (Bolongaro-Crevenna-Recaséns et al. 2010). Three substratum layers have been defined on the beach: a superficial layer (1–30-cm depth) of light, low-humidity sandy grains; an intermediate layer (30–180-cm depth) of high-humidity grains; and a bottom layer (> 2-m depth) molded by coarse biogenic conch remains (Bolongaro-Crevenna-Recaséns et al. 2010).

View Figure

• Download as Powerpoint
• Download Figure

During an annual cycle, the region experiences 3 different climatic seasons: dry (March–June), rainy (June–October), and norte (sequential storms out of the north; November–February) (Álvarez and Barrios 1999). Dune vegetation on Chenkan Beach forms a fringe of varying width along the beach's inland side. It consists mainly of bushy and creeping species such as seagrape (Coccoloba uvifera), siricote (Cordia sebestana), bayhops (Ipomea pes-caprae), jacquinia (Jacquinia flammea), and coconut (Coccus nucifera) (Bravo 1985; Reyes-Gómez and Vásquez-Lule 2009). Fauna on the beach and in dune zones include ghost crabs (Ocypode sp.), raccoons (Procyon lotor), white-nosed coati (Nasua narica), gray fox (Urocyon cinereoargenteus), green iguana (Iguana iguana), black iguana (Ctenosaura similis), anteater (Myrmecophaga sp.), opossum (Didelphis marsupialis), common long-nosed armadillo (Dasypus novemcinctus), pelican (Pelecanus occidentalis), common frigate (Fregata magnificens), gulls (Laurus spp.), and boa (Boa constrictor) (Escanero-Figueroa et al. 1990). The marine habitats neighboring the beach are shallow (< 5 m) and covered mainly by seagrass and algae beds that support a variety of sponges, echinoderms, crustaceans, and mollusks.

Nesting Activity

Biological monitoring of turtle nesting activity on this beach is done by personnel of the Laguna de Términos Flora and Fauna Protection Area of the National Commission of Natural Protected Areas (Consejo Nacional de Areas Naturales Protegidos [CONANP]). Night patrols (from about 2100 to 0600 hrs the next day) are done throughout the turtle nesting season (May–October) during which personnel search for nests and nesting females. Confirmed nests are recorded every night, and a set of biological and physical data are collected including the geographic coordinates of each nest. After the incubation period, 2 biological parameters associated with incubation success were calculated: 1) hatching success, expressed as the percentage of hatchling turtles that broke through the eggshell; and 2) emergence success, which is the percentage of hatchlings that left the nest unaided (this is estimated by subtracting the number of live and dead hatchlings remaining in the nest from the total number of recorded eggs).

Environmental Variables

Beach slope and width were measured at 15 sites along the nesting beach with approximately 700–2000 m between each measured nest. Beach profile was characterized for 3 zones: intertidal, berm, and dune (Guzmán and García 2010). Profile data were recorded with a hand level (CST model 17621) for slope and a conventional plastic tape measure for width (50 m in length). The beach profile data were used to calculate slope in degrees using the following equation (Márquez-García 2002):

where DV is vertical distance and DH is horizontal distance.

For each nest, measurements were taken of its depth, its distance to the highest tidemark (based on the highest location of algae washed up on the beach), and the shortest distance to vegetation. Vegetation community data were collected by photographing vegetation cover in a 1.5-m² area surrounding the nest, focusing the camera from the top of the nest at a distance of 1.8 m. Each photo was analyzed using the Coral Point Count CPCe V3.5 tool to calculate vegetation cover per square meter (Kohler and Gill 2006). Counts were done per nest, and percentages calculated for each vegetation type. Recorded vegetation was identified to the species level using illustrated guides for Yucatan Peninsula Flora (Guadarrama et al. 2014). This approach only evaluates the effect of the vegetation cover in the area immediately around an incubating nest; the temporal dynamics of vegetation cover exceed the present study scope, although its potential effect warrants further research.

Temperature data loggers (UA-001-08 Hobo pendant, Onset Computer Corporation) were used to record nest internal temperature. Sensors were placed in the central portion of the nests during egg laying. The loggers were programmed to take temperature readings every hour during the entire incubation period (approximately 55 d). Complete temperature measurements were obtained from 17 nests, which corresponded to the number of available loggers. Nests that predated logger placement were excluded from the analysis.

Statistical Analyses

Linear regression analyses were run to assess the relationship between beach morphology (width and slope) and the number of hawksbill nests (reproductive parameters). Data for the entire nesting beach was used; each analyzed site produced a description of spatial variation in physical characteristics. Partial regressions were also done for 3 separate months (May, July, and September), considered representative of the 3 regional climatic seasons, to evaluate whether beach width and slope influenced the number of nests.

A redundancy analysis (RDA) was used to analyze how hatching success (HATsuc) and emergence success (EMEsuc) correlated to the environmental variables measured on all recorded nests. Graphic ordination plots were produced using 2 different scaling methods: scaling 1 to enhance the relationships between biological parameter response variables, and scaling 2 to enhance the correlations between the independent environmental variables. These diagrams helped to identify the most-important environmental variables affecting turtle reproductive parameters.

Twelve variables were considered in the redundancy analyses: 1) slope of the intertidal beach sector (slope A); 2) slope of the beach berm (slope B); 2) slope of the beach segment between the berm and vegetation line (slope C); 4) total slope (average of the three independent slope measurements); 5) distance from nest to the high-tide mark; 6) distance from nest to nearest vegetation; 7) nest depth; 8) beach width (measured from high-tide mark to dune reference points, i.e., wooden stakes or permanent points established at onset of monitoring and used throughout sampling period); 9) creeping vegetation cover; 10) herbaceous vegetation cover; 11) bushy vegetation cover; and 12) arboreal vegetation cover. All analyses were done using the statistics package Vegan (Oksanen et al. 2013) for the R numerical environment (R Development Core Team 2013).

RESULTS

Nesting Activity

A total of 78 hawksbill turtle nests were recorded along the 18 km of Chenkan Beach from April to October 2014. Mean hatching success was 66% (SD = 25.4) whereas emergence success was 60% (SD = 26.8).

Beach Physical Parameters and Nest Distribution

Beach slope ranged from 4° to 12° (x̄ = 6.99, SD = 2.04) in the 15 studied sections. Beach width ranged from 12 to 28 m (x̄ = 21.62, SD = 6.92). The steepest slope and shortest width values occurred in the beach's middle section (section 7) while the gentlest slope and largest width were in the eastern extreme of the study area (section 15) (Fig. 2). Nest spatial distribution did not significantly correlate to 2 physical parameters (Table 1). The regression analysis results showed no significant relationship between beach slope or width and the number of nests during the 2014 nesting season for the overall length of beach (Table 1; Fig. 3).

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Results of the regression analysis between number of clutches and beach physical environment variables. r² = coefficient of determination, m = slope, and p = statistical probability value.

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

Relationship Between Habitat Variables and Turtle Reproductive Parameters

The plant community associated with the hawksbill nests included 11 species, all typical of coastal dunes on the Yucatan Peninsula. Total surface area data for the nests showed 50% to have no vegetation and the rest to be covered by creeping, herbaceous, bushy, and arboreal vegetation (or some combination) (Table 2). The distance from the nests to the nearest vegetation ranged from 0.1 to 2.5 m (x̄ = 0.46 m, SD = 0.35). Nest distance to the highest tide mark ranged from 4.3 to 48.8 m (x̄ = 15.6 m, SD = 6.98). Nest depth ranged from 0.41 to 0.67 m (x̄ = 0.51 m, SD = 0.06).

Table 2. Plant species, type of structural vegetation, and mean percentage associated with turtle nests within a 1.5-m² area surrounding each nest. NV = no vegetation.

View Fullscreen

The ordination analysis showed that the main ordination axis runs approximately parallel to the first axis (RDA1), distinguishing 2 groups: one on the left side containing most of the nests, and one on the right side containing fewer nests (Fig. 4). A high proportion of the total variation in the response variables (HATsuc and EMEsuc) was clearly associated with the group containing the most nests (left side). All the analyzed environmental variables explained 49% of total variation in the measured reproductive parameters; an analysis of variance (ANOVA) permutation test showed this to be significant (p < 0.05). The habitat variables that contributed to explaining high variation in hatching and emergence success included nest depth (NestDep), distance to vegetation (DisVeg), and coverage of beach substratum by different vegetation types, e.g., bushy (BushVeg), arborous (ArboVeg), and herbaceous (HerbVeg) plants but not creeping plants (CreeVeg). Creeping plants apparently compete with other vegetation types and were inversely related to turtle reproductive success and other vegetation types (Fig. 4).

View Figure

• Download as Powerpoint
• Download Figure

Nest Thermal Regime

Average temperature in the 17-nest sample was 31.78°C (SD = 0.74) and average vegetation cover area surrounding the nest was 61%. The presence of vegetation around the nest influenced nest temperature; for instance, Nest 7 had no surrounding vegetation and an average temperature of 31.97°C, whereas Nest 6 had 100% vegetation cover and a temperature of 29.47°C, the lowest temperature of all evaluated nests (Table 3).

Table 3. Mean temperature and percentage of vegetation and reproductive parameters measured for 17 turtle nests on Chenkan Beach, Campeche, Mexico.

View Fullscreen

The highest temperature was recorded in Nest 11 (33.05°C), which had 29% vegetation cover dominated by creeping vegetation. The same was the case for Nests 8, 9, 12, and 16, where temperature was higher than the mean temperature for all nests (Table 3). In contrast, nests covered by herbaceous and shrub type vegetation had a lower temperature than the mean average (e.g., Nests 1, 2, 5, 6). Only Nest 4 was adjacent to trees and its temperature was 30.46°C, below the mean temperature. Hatching and emergence success was highest in Nest 4 and lowest in Nests 11 and 12, both covered by creeping vegetation. Of note is that Nest 10 was a relocated nest. The female had made her nest in the intertidal zone, leaving her eggs at risk of flooding. We relocated the nest to a dune area dominated by creeping and herbaceous vegetation, where its average temperature was 32.07°C, with a 95.8% hatching success and a 56.9% emergence success (Table 3).

DISCUSSION

Hawksbill turtle females showed no nesting preferences associated with beach slope and width. Previous studies have demonstrated that narrow beaches are vulnerable to sea level rise, forcing turtles to modify nesting site selection due to limited suitable habitat (Fish et al. 2005; Bolongaro-Crevenna-Recaséns et al. 2010). Turtles may be able to adapt to small-scale changes in beach characteristics over short time periods (Hamann et al. 2007; Hawkes et al. 2009), so it is important to accommodate conservation planning for hawksbill turtle nesting to local nesting beach characteristics (Liles et al. 2015).

The lack of a relationship between absolute beach slope and width and the number of nests suggests that turtles may be responding to changes in local beach morphology dynamics (Fuentes et al. 2013). Some studies have shown that hawksbills preferentially nest on steep beaches in the Caribbean Sea (Fish et al. 2005) while others suggest they nest at a certain elevation above sea level (Horrocks and Scott 1991). The present findings coincide with Cuevas et al. (2010), who suggested that hawksbills exhibit plasticity in nesting site preference in response to frequent changes in local beach morphology, as observed at Chenkan Beach.

Turtle fidelity to particular microhabitats changes in response to patterns in biological processes, and the main structural components of such microhabitats include vegetation type and beach width (Kamel and Mrosovsky 2006b). Of the habitat variables measured in the present study, nest depth and distance to adjacent vegetation (particularly shrubby, herbaceous, and arboreal plants) positively affected hatching and emergence success. Unlike other marine turtle species, the hawksbill tends to select nesting sites with particular vegetation characteristics (Kamel 2013). Kolbe and Janzen (2002) and Kamel and Mrosovsky (2005) showed that hawksbill turtles primarily nest in heterogeneous habitats composed of a patchwork arrangement of areas with and without vegetation. Other studies indicate that pioneer herbaceous vegetation species have a consistent effect between nesting seasons on hawksbill turtle nest site choice along with other variables such as the vegetation line, distance of nest from current waterline, and the highest spring tide (Santos et al. 2016).

The hawksbill population nesting on Chenkan Beach primarily nested at sites containing vegetation, supporting a previous study on other beaches in the state of Campeche which reported that more that 60% of nests were shaded by vegetation (Guzmán et al. 1995). The present results suggest that vegetation, mainly trees and shrubs, regulates the temperature around nests, maintaining it at levels that facilitate hatching and emergence. It is thus clearly important to maintain natural vegetation structure, type, and coverage patterns to optimize hawksbill turtle nesting habitat.

The inverse relationship between creeping vegetation and the other vegetation types shown in the ordination diagram is explained by the effect plants had on variation in reproductive variables, i.e., creeping vegetation produced a negative effect on variation in hatching and emergence success compared with the positive effect of the other vegetation types. This agrees with previous studies indicating the presence of creeping plants to be negatively correlated to nest depth (e.g., Kamel 2013), perhaps due to the presence of roots, which prevent females from forming adequate nests and thus slows egg development. This highlights the importance of carefully selecting plant species when restoring dune habitat on nesting beaches. Native species are generally preferred when restoring dune habitat (Moreno-Casasola et al. 2008), but any plant species that potentially facilitate nest temperature regulation should be considered. Future studies should determine if the presence of creeping plants promotes the expansion of other vegetation types through succession which could, in turn, increase the number of turtle nests and improve reproductive output.

The incubation success variables of hatching and emergence success were higher in dune zones with vegetation in which nests had lower mean temperatures. A negative relationship was observed between the presence of creeping vegetation and hatching and emergence success. This supports previous reports that creeping plant species such as Ipomoea pes-caprae reduce hatching success because roots enter the nest, often harming the eggs (Conrad et al. 2011), trapping emerging hatchlings, or both.

The present results coincide with the widely accepted observation that vegetation is an important beach structural component influencing the hawksbill turtle nesting process. Hatching success was higher in nests surrounded by vegetation, which supports the findings of Kamel (2013) but contrasts with those of Ditmer and Stapleton (2012). In addition to the spatial pattern of nest arrangement on beaches, the percentage of plant coverage and vegetation type are both important habitat factors that contribute to successful hawksbill turtle reproductive output. Novel visual techniques might reveal new insights into how the spatial pattern of vegetation growing on beaches influences the nesting process and reproductive success of marine turtles.

Overall, hawksbill turtles did not show preferences for particular nesting beach slopes and widths, suggesting they may be flexible users of different sites if severe changes in beach morphology occur. For instance, turtles were observed to avoid a beach sector where breakwater structures had been installed, meaning turtles that had consistently nested in this location were forced to move to alternative sites. Breakwater structures are particularly controversial because they reduce beach erosion in the immediate vicinity but may represent an obstacle for transit of female turtles to the beach.

Vegetation was a central structural component of hawksbill nesting habitat, far more so than beach physical characteristics. Vegetation limits beach erosion and degradation as well as preventing temperature increases in nests. Hatching and emergence success were generally higher in sectors where the dune zone had high vegetation coverage. The present findings indicate that vegetation restoration programs on turtle nesting beaches should consider vegetation spatial distribution patterns and the vegetation type present to enhance hatching production on beaches where high temperatures are an issue; these can compromise embryo survival and lead to highly female-biased sex ratios. This issue could become more salient in the coming decades as climate change raises overall global temperatures.