The Role of Sand Moisture in Shaping Loggerhead Sea Turtle (Caretta caretta) Neonate Growth in Southeast Florida
Abstract
Many environmental variables that affect incubating turtle eggs in the nest may also affect hatchling development, following hatchling emergence. However, these effects may be subtle and are largely unexamined. In this study, we analyzed the effect of sand moisture content during incubation on the postemergence growth rates of loggerhead sea turtles (Caretta caretta) in southeastern Florida. We divided 10 clutches in halves, reburied them, and exposed them to 1 of 2 treatments. At emergence, 7 clutches met minimum criteria for inclusion in the study. One half-clutch received only ambient rainfall (“dry” treatment) while the other half-clutch received ambient rainfall plus daily watering (“wet” treatment). Data loggers were used to record incubation temperatures in both groups. Hatchlings were captured at emergence and laboratory-reared over a period of ∼ 3 mo. Mass, straight carapace length (SCL), and straight carapace width (SCW) were measured weekly to track growth. Initial measurements were larger for turtles from the wet nests in all metrics. Turtles from wet nests grew more in SCW than turtles from dry nests. Turtle growth from the 2 treatments did not differ in SCL or mass measurements. Larger initial sizes and faster SCW growth may enable the turtles to more quickly achieve a refuge size from their gape-limited predators. Moisture availability during nesting season is projected to decrease based on climate change models. If that change materializes, it could negatively affect hatchling sizes and neonate growth rates, survival, and hence the recovery of this imperiled species.
During development, sea turtle embryos are influenced by environmental factors such as temperature, gas concentrations, and water vapor within the surrounding beach sand. For example, nest temperatures affect various aspects of embryonic development including growth and physiological rates, and hatchling size. High temperatures and drier conditions are associated with scute anomalies (Herlands et al. 2004; Reid et al. 2009; Telemeco et al. 2013; Kobayashi et al. 2017; Zimm et al. 2017) and an increase in embryo mortality (Ackerman 1997; Matsuzawa et al. 2002). Nest temperatures also influence sex ratios because all sea turtle species exhibit temperature-dependent sex determination (TSD). Cooler temperatures result in an increase in the proportion of males, and warmer temperatures increase the proportion of females (Yntema and Mrosovsky 1980; Miller and Limpus 1981; Standora and Spotila 1985). The metabolic activity of the developing embryo is supported by the exchange of gases across the permeable eggshell. If gas conductance is limited (e.g., by fine-grained sands), incubation duration is increased while both hatching success and embryonic growth rates are decreased (Ackerman 1997).
In addition, turtle eggs exchange moisture with the surrounding environment in the form of water vapor and possibly liquid water. Increased moisture from rainfall can affect embryonic growth rate, incubation duration, and hatchling size at emergence. Sufficient moisture is essential for normal development. Previous studies of freshwater and marine turtle species suggest that the hydric environment may affect various aspects of embryonic development including incubation duration (Steyermark 1999), amount of residual yolk remaining at hatching (Packard et al. 1983, 1987), and hatchling size (Paukstis et al. 1984; Packard et al. 1989; Brooks et al. 1991; Cagle et al. 1993; Spotila et al. 1994). Tucker et al. (1998) found that even relatively small changes in moisture produced measurable effects on embryonic development, specifically hatchling size, residual yolk size, and incubation period. Proper moisture content of nest substrate is also crucial for sea turtle hatching success (McGehee 1990). Both dry and extremely wet nest substrates are associated with decreased hatching success. Nest sand moisture increases with rainfall or the nest's proximity to the tide line (Carthy et al. 2003).
A number of studies suggest that the hydric environment, in combination with the thermal environment, influences sex in sea turtles as well (e.g., LeBlanc and Wibbels 2009; Wyneken and Lolavar 2015). However, with the exception of sex ratio studies, the influence of environmental factors during incubation on postemergence development of neonate sea turtles is largely unknown. In this article, we explore the influence of differences in moisture during incubation on neonate growth during the first few months after hatching. We examine how increased sand moisture during incubation affects 1) initial hatchling size and 2) turtle growth over a period of ∼ 3 mo.
METHODS
Study Site and Nests
Eggs were obtained from 10 loggerhead (Caretta caretta) nests, which were relocated because they were deposited too close to the surf zone on the beach at Boca Raton, Florida (26°22′09.5″N, 80°04′04.3″W). Nests were excavated the morning after deposition. All the eggs were placed in moist beach sand contained in small, insulated boxes and transported to our study site within 10–15 min. The study site was located ∼ 30 m landward from the high tide line. This location was typical of sites selected by nesting females and received some natural nests every year. Each clutch was divided into 2 smaller half-clutches of equal or approximately equal size. The half-clutches (hereafter termed nests) were reburied in standardized nest holes with a narrower neck (16–21 cm in diameter) and wider bottom (23–26 cm in diameter, ∼ 60 cm in depth to the bottom of chamber) based on Miller et al. (2003).
Treatments
The clutch pair served as a control (“dry”) group and an experimental (“wet”) group. All nests within a treatment were ∼ 1 m apart. A past study (Lolavar and Wyneken 2015) found that rainfall events of ∼ 3 cm resulted in decreases in nest temperature, which was interpreted as an effect of increased nest moisture. Thus, wet nests received daily watering with clean freshwater from the city water supply for ∼ 45 min (3–5 cm) to ensure that water treatments reached nest depth and did not dissipate before reaching the egg mass. The water was distributed by an oscillating sprinkler, which had a span of ∼ 8 m and a spray area of ∼ 27 m2. Dry nests were away from the sprinkler zone, ∼ 5 m seaward of wet nests, and ∼ 40 m from the surf. Natural rainfall also supplied water to both groups.
Temperature and Moisture Measurements
Temperature data loggers (HOBO U22-001: accuracy ± 0.21°C, resolution 0.02°C per manufacturer specifications; Onset Computer Corp, Bourne, MA) were placed in the center of every half-clutch (the egg mass) and recorded data every 15 min through incubation. The sand surrounding each half-clutch was equipped with a S-SMC-M005 Soil Moisture Smart Sensor (accuracy ± 0.031 m3/m3, resolution 0.0007 m3/m3; Onset Computer Corp) to measure nest sand moisture content at 15-min intervals.
Turtle Rearing and Measurements
After nests reached 45 d of incubation, we placed a cage formed from hardware cloth (wire mesh) on top of the egg chamber. Cages were closed every night and checked periodically until sunrise the following morning for hatchling emergence. If no hatchlings were observed, cages were opened so any hatchlings that emerged during the day could crawl to the ocean. Once hatchlings emerged, up to 10 hatchlings were collected from each nest and immediately transported to our nearby marine laboratory. If a nest produced < 10 hatchlings, either due to a missed emergence or low hatch success, all hatchlings from that nest were collected. Hatchlings were assigned an ID number, marked with nontoxic nail polish for identification, measured using Vernier calipers (accuracy, 0.1 mm; straight carapace length [SCL], straight carapace width [SCW]), and weighed (in g, using an electronic scale [OHaus Scout Pro 601, OHaus Corporation, Parsipanny, NJ; accuracy, ± 0.1 g]). Growth was tracked by repeating these measurements weekly. All turtles were placed in quarantine tanks and joined the study if they swam, fed, and defecated normally.
All turtles were housed individually, in flow-through baskets floating at the surface, in tanks continuously supplied with fresh seawater. Ambient ocean temperatures ranged between 25°C and 29°C during the 8–12 wks that the turtles were reared, and all pairs of dry and wet nests experienced the same water temperatures during the same time period. All turtles were fed a standard in-house manufactured diet (Stokes et al. 2006) at 11% of body weight/day.
Calculations and Analyses
All measurements were tested for normality using Shapiro-Wilk tests. Temperature has a known relationship with embryonic development and hatchling growth rates (Ackerman 1994; Booth et al. 2004; Lolavar and Wyneken 2015). Incubation temperatures also were compared between the wet and dry treatments because of the potential for evaporative cooling by added moisture. A mean temperature at each 15-min interval was calculated for each treatment and the temperature profiles were compared using a 2-tailed t-test. This allowed us to identify a potential influence of temperature on observed differences in growth. Moisture levels between groups were compared by averaging moisture values at each time interval and comparing moisture profiles using a 2-tailed t-test. Initial SCL, SCW, and mass were compared between treatments using a univariate analysis of variance (ANOVA) with clutch as a random factor. Repeated-measures analysis of covariance (ANCOVA) compared measurements (SCL, SCW, and mass) over 10 wks between the wet and dry treatments with initial hatchling size as a covariate, as suggested by Packard and Boardman (1999).
An α = 0.05 was used to reject the null hypotheses that moisture had no effect on neonate growth, nest temperatures, and nest moisture.
RESULTS
In our study, sprinkling water decreased nest temperatures by 0.2°–0.5°C and increased moisture by ∼ 0.015 m3/m3 compared with ambient nests. Mean nest temperatures differed significantly between the wet and dry nests; wet nests were cooler than dry nests (Fig. 1). Mean nest moisture also differed significantly between the treatments, with wet nests having higher average moisture levels (wet = 0.068 ± 0.027 m3/m3 SD; dry = 0.053 ± 0.026 m3/m3 SD; t2,7994 = 100.45; p < 0.0001).
Citation: Chelonian Conservation and Biology 17, 2; 10.2744/CCB-1301.1
All nests that were undisturbed produced hatchlings. Some nests experienced predation and low hatching success and 6 turtles failed to pass quarantine; therefore, samples were not always equal between nest pairs. At emergence, 7 pairs of clutches met minimum criteria for inclusion in this study. Nest pairs with ≤ 5 hatchlings in either treatment were not included in the study. Seven nest pairs (14 half-clutches) produced 122 hatchlings (67 wet, 55 dry) that were included in this study. Shapiro-Wilk tests for normality found that we could not reject the hypothesis that the samples come from populations with a normal distribution, for most samples (Table 1). Zar (1999) states that sample sizes > 50 are large enough to provide a robust analysis even with a departure from normality. Therefore, we were able to use a univariate ANOVA with clutch of origin as a random factor for comparisons of our samples because our sample sizes are large (67 wet, 55 dry) and the departures from normality were small. This ANOVA showed us that hatchlings from wet nests were initially significantly larger than the hatchlings from dry nests in all metrics (Table 1; Fig. 2).
Citation: Chelonian Conservation and Biology 17, 2; 10.2744/CCB-1301.1
We compared growth in all metrics while accounting for initial hatchling size as a covariate. Wet treatment nests produced hatchlings that grew significantly more in SCW in the months after emergence compared with turtles from the drier control nests (Table 2; Fig. 2). When we accounted for initial hatchling size, growth in SCL and mass did not differ significantly.
DISCUSSION
Nests that received more water were cooler and moister than nests that received only ambient rainfall. Hatchlings from those wet nests initially were larger than their dry nest counterparts in all metrics, SCL, SCW, and mass. Studies of freshwater turtles show that eggs incubating in drier conditions produce smaller hatchlings with larger yolk reserves and larger hearts (presumably in response to more viscous blood; Packard et al. 1988; Packard and Packard 2001). Morris et al. (1983) found that snapping turtle eggs incubating in wetter conditions took up water, and the embryos' utilization of the yolk energy reserves were greater later in development than turtles that incubated under drier conditions. Our loggerhead hatchlings from wet nests hatched at a larger average size. We infer these hatchlings had less yolk reserves than their smaller siblings from drier nests, because larger size requires more yolk utilization and they incubated (and grew) longer than their clutch-mates in the dry treatment.
Wet treatment hatchlings grew significantly more in SCW than dry treatment hatchlings. These results do not support the null hypothesis that moisture conditions during incubation have no prolonged effects on neonate growth. By dividing single nests into 2 treatments that were incubated in close proximity, we were able to control for potential maternal influences on eggs, control temporal and spatial beach variation, and verify this prolonged effect of moisture.
Our study adjusted moisture content of wet treatment nests with daily watering and the added moisture slightly reduced nest temperatures. These results have direct implications to the effects of moisture on naturally incubating nests. This study aimed to examine the impact of increased nest moisture on growth. Past studies suggest that increases in nest moisture by rain are likely to cause a decrease in nest temperatures (Houghton et al. 2007; Lolavar and Wyneken 2015). Matsuzawa et al. (2002) examined nests in Japan and found sand temperature increased as the rainy season ended; thus, moisture influences the thermal environment of the nest. Naro-Maciel et al. (1999) reported that 1–3 hrs of sprinkling water on nests decreased nest temperatures by ∼ 1°C. In our study, sprinkling decreased temperatures by 0.2°–0.5°C and increased moisture by ∼ 0.015 m3/m3 compared with ambient nests. Changes in thermal environment can affect embryonic development because incubation temperatures affect incubation duration, embryonic physiology, hatchling size, and embryo survival (Ackerman 1997). Jourdan and Fuentes (2015) suggest that manually wetting nests may not reduce sand temperatures in hot areas. The 2015 nesting season, along southeastern Florida, was characterized by record-breaking high temperatures and severe drought during the first half of incubation (Heim 2015; National Oceanic and Atmospheric Administration National Centers for Environmental Information 2015). The effect of watering nests in our study was to significantly decrease nest temperatures, even during this exceptionally hot year. The combined effects of temperature and moisture affected initial hatchling size. Past studies found similar results when examining nest moisture's effect on initial hatchling size (McGehee 1990 [C. caretta]; Finkler 2006 [Chelydra serpentina]). McGehee (1990) found that only SCL increased significantly with higher sand moisture levels; our study found differences in initial SCL, SCW, and mass.
Our results have direct implications for characterizing hatchling condition at Florida's east coast loggerhead rookery, one of the largest loggerhead rookeries in the world. From this research, we are able to confirm that moisture during incubation is associated with initial size differences, and greater growth of an important metric (SCW) in loggerhead hatchlings. Available moisture may influence hatchling size and growth by creating a more favorable environment for the embryo to convert yolk into tissue (i.e., muscle, fat, skeleton).
Getting larger sooner means that neonates gain size refuge from common predation threats (Gyuris 1994; Salmon and Scholl 2014). Similar growth patterns have been seen in the red-eared slider (Trachemys scripta), a freshwater species (Janzen et al. 2000). Wet nest hatchlings immediately gain a size advantage resulting from their larger initial size and can maintain this size advantage as a result of higher growth in SCW. Ultimately, hatchlings from nests with more rainfall may experience reduced mortality from gape-limited predators given their initial larger size and continued higher growth. Increased neonate survival can have positive down-stream implications for a sea turtle species that is globally listed as vulnerable (Casale and Tucker 2017). On the other hand, if drought conditions persist, hatchlings are likely to hatch at a smaller size (McGehee 1990) and grow more slowly, thus leaving them vulnerable to gape-limited predators for a longer period of time.
The impacts of climate change in Florida will vary by location within the state and season (Florida Oceans and Coastal Council 2010; Misra et al. 2011). In the 21st century, peninsular Florida is predicted to experience overall drier conditions during June, July, and August, which coincides with peak loggerhead nesting. Climate models for Florida project an 8% decrease in average rain during these months (Misra et al. 2011). Southern Florida likely will experience drier conditions compared with northern Florida. Periods of drought are expected to be punctuated by increased frequency of heavy rainfall from intense storms (Goldenberg et al. 2001; Webster et al. 2005; Misra et al. 2011; Intergovernmental Panel on Climate Change 2014). Thus, nests may experience extremes in moisture: mostly dry conditions along with sudden large amounts of rain in a short span of time. Extremes may be detrimental because the parchment-shelled eggs of sea turtles and the developing embryos can regulate water balance within normal ranges. Consequently, for imperiled species, understanding moisture's role is an essential aspect of assessing possible impacts on effective hatchling recruitment. Although moderate and consistent moisture during incubation is necessary for developing embryos, sudden excessive rain results in egg suffocation or nest washout (Kraemer and Bell 1980). Heavy rainfall decreases gas diffusion throughout the nest (Miller et al. 2003), which can cause egg death, if extreme. Together, climatic shifts and extreme weather events can have direct negative impacts on hatchling production and on the proportion of neonates surviving and recruiting to the next size classes. Understanding that this decrease in moisture during incubation may adversely affect hatchling growth rate is essential in assessing hatchling recruitment.
Mean temperatures (°C) for wet (32.30 ± 0.96 SD, n = 10, gray) and dry (32.46 ± 0.94 SD, n = 9, black) loggerhead sea turtle nests. Means represent temperatures taken every 15 min during incubation. Two-tailed t-test shows that dry nests experienced significantly higher temperatures (t2,8971 = 12.682; p < 0.0001).
Neonate loggerhead sea turtle growth over 10 wks for mass, straight carapace length (SCL), and straight carapace width (SCW). Each graph shows wet (gray) vs. dry (black) turtle size. Error bars represent 95% CI. Wet treatment hatchlings were initially larger in all metrics (mass, SCL, and SCW; Table 1). Only SCW increased significantly more for the wet treatment turtles over the 10-wk period (Table 2).
Contributor Notes
Handling Editor: Jeffrey A. Seminoff
