Response of Nesting Sea Turtles to Barrier Island Dynamics
ABSTRACT
Although barrier island beaches provide important nesting habitat for sea turtles, they are constantly changing. To determine how nesting sea turtles have responded to this dynamic environment, we assessed: 1) wind, current, and tidal patterns and changes in beach profiles, 2) sea turtle nesting patterns, and 3) success of turtle nests deposited along 5 km of beach on Cape San Blas, Florida, an extremely dynamic barrier beach in northwest Florida. From 1998 to 2000, nesting turtles were tagged, nests were monitored, and hatching success was determined. Throughout this study, West beach lost ∼ 5 m of sand while East beach gained ∼ 4 m; however 61% of nests were deposited on West beach and 39% on East beach. Hatchling emergence success did not differ between beaches. Wind direction influenced current direction and sand movement and affected the number of nests deposited along East beach but not on West beach. Nearly all nests (98%) oviposited on both beaches were deposited during a rising tide. Although West beach is narrow and eroding, the steep slope may enable nesting turtles to expend less energy to reach higher nesting sites while still providing successful nests. Nesting on a rising tide and using offshore currents during the internesting period may assist this effort.
Female loggerhead turtles (Caretta caretta) nest every 1 to 3 years, and from 1 to 6 times within each reproductive season (Miller 1997). It has been suggested that female turtles return to their natal beach to nest, and that once a female has returned to the region of her birth she will tend to renest in close proximity on subsequent nesting events within and between nesting seasons (Carr and Hirth 1962; Carr and Carr 1972; Talbert et al. 1980; Williams-Wallis et al. 1983). Genetic and tagging studies support this theory (Carr 1975; Pritchard 1976; Bowen et al. 1992; Encalada et al. 1996). Changes in the morphology of a nesting beach due to ocean currents, winds, and tides may present challenges to turtles attempting to return to their natal nesting beach. Effects of these forces on nest site selection by loggerhead turtles are largely unknown.
When attempting to nest, loggerhead turtles must first select a beach, then emerge from the water, and finally deposit the clutch on that beach (Wood and Bjorndal 2000). Beach characteristics such as temperature, salinity, slope, moisture, width, and sand type have been shown to influence nest placement (Johannes and Rimmer 1984; Garmenstani et al. 2000; Wood and Bjorndal 2000). When optimal, these factors may allow turtles to expend less energy in locating nesting sites that will provide the greatest reproductive success. Along dynamic beaches, these factors are constantly changing, which may reduce a turtle's ability to identify high-quality nesting sites.
Barrier island beaches typically undergo severe erosion and accretion throughout the year; however, these habitats are also often used by nesting loggerhead turtles. Along the eastern coast of the United States, loggerhead turtles nest on several barrier islands, including Topsail Island, North Carolina (Grant and Beasley 1998); Kiawah Island, South Carolina (Talbert et al. 1980); Little Cumberland Island, Georgia (Frazer 1983b); and Hutchinson Island, Florida (Williams-Wallis et al. 1983). In addition, barrier islands form almost half of the Gulf of Mexico shoreline, and loggerhead turtles commonly nest in this region (LaRoe 1976; LeBuff 1990). The dynamic habitat along these barrier islands may provide significant challenges for nesting loggerhead turtles.
Along the northern Gulf of Mexico the greatest density of loggerhead turtle nesting occurs along 5 km of beach owned by the US Air Force on Cape San Blas, Gulf County, Florida. This barrier beach is located along the Florida panhandle and represents the southernmost point of the St Joseph Peninsula (Fig. 1). From 1993 to 1997 this area supported a mean of 48 loggerhead turtle nests/yr (9.5 nests/km; Meylan et al. 1995; Lamont et al. 1997). No other species of sea turtle has been documented nesting at this site.
Citation: Chelonian Conservation and Biology 6, 2; 10.2744/1071-8443(2007)6[206:RONSTT]2.0.CO;2
Cape San Blas supports a significant group of nesting turtles and genetic studies have indicated turtles from this region represent a unique stock (Encalada et al. 1998). However, this site also hosts extremely dynamic beaches. The eastern (south-facing) beach (hereafter referred to as East beach) of Cape San Blas undergoes accretion, whereas the western coast (hereafter referred to as West beach) experiences some of the greatest erosion rates in Florida. From June 1994 to September 1995, approximately 10 m of sediment was eroded from West beach (Lamont et al. 1997). Although West beach is less stable than East beach, sea turtles using Cape San Blas tend to nest along the eroding rather than the accreting beach. From 1994 through 1997, at least 60% of nests deposited on Cape San Blas were deposited on West beach (Lamont et al. 1997). How the dynamics of this environment influence nesting activity of loggerhead turtles is unknown.
To determine how nesting sea turtles have responded to this dynamic environment, our goals were to 1) assess dynamics of Cape San Blas, including wind, currents, and tidal patterns and changes in beach profiles, 2) determine sea turtle nesting patterns, and 3) quantify hatching success of nests deposited along Cape San Blas.
METHODS
Sea Turtle Surveys and Reproductive Success
Daily morning surveys for sea turtle nests were conducted from 15 May through 15 September in 1998, 1999, and 2000. Nests were marked with 4 wooden stakes wrapped with orange flagging, placed around the body pit. Starting at 45 days of incubation, nests were visually examined every morning for signs of hatching. One week after the last hatchling emergence or after 90 days incubation, nest excavations were conducted to evaluate success. In addition, night surveys were conducted from approximately 2100 to 0600 hours every night during the nesting season (15 May to 10 August). When a nesting turtle was located, we identified the species, measured curved carapace length (CCL; from nuchal notch to the longest projection of the pygal) and curved carapace width (CCW; edge of carapace to edge of carapace in the widest region), and tagged each turtle with Inconel flipper tags (National Band and Tag Company, Newport, KY) placed in the trailing edge of both front flippers. The location of each nest was recorded using a Global Positioning System (GPS). Nests deposited below mean high water were relocated landward or to a more stable location. For analysis, sea turtle nests deposited west of the cape spit (between mile markers 1.4 and 2.9) were categorized as being deposited on West beach and those deposited east of the cape spit (between mile markers 0.0 and 1.4) were categorized as being deposited on East beach. For correlations with tidal height, we used time of the nesting female's emergence or the time when a female was first observed, rather than time of egg deposition.
Success was defined as the number of hatchlings that emerged from the nest divided by the total number of eggs deposited in the nest, and was termed hatchling emergence success following Johnson et al. (1996). These calculations included nests lost to erosion or depredation. Because we were comparing success among geographic regions we wanted to ensure that hatchling success reflected the area where the nest was deposited, therefore nests that were relocated were not included in calculations of hatchling emergence success. The total number of eggs in the nest was assessed during nest excavation and was determined by adding the number of hatched eggs (all eggshells representing greater than 50% of a whole egg), unhatched eggs, and pipped eggs. To calculate the number of hatchlings that emerged from the nest, the number of dead hatchlings found within the nest was subtracted from the total number of hatched eggs. Eggs that contained developed hatchlings that had not pipped or emerged from the egg were considered unhatched eggs.
A Student t-test was used to test for significant differences in the number of nests deposited between locations (Zar 1984). A Student t-test or a nonparametric Mann-Whitney Sum Rank test was used to test for significant differences in hatchling emergence success and number of nests lost to erosion between east and west beaches (Zar 1984). The nonparametric Mann-Whitney Sum Rank test was used when assumptions of normality (Kolmogorov-Smirnov test) or equal variances (Levene median test) were not met (Zar 1984). All statistical analyses were performed using SigmaStat 2.0 (Jandel Corporation 1995) unless otherwise noted.
The relatively small sample size of this study often made the use of nonparametric statistics necessary. Nonparametric statistics do not depend on the assumptions that the samples collected are from populations that have normal distributions and equal variances however nonparametric statistics are also generally not as powerful as parametric statistics (Zar 1984). The nonparametric test used for analysis of these data was the Mann-Whitney Sum Rank Test. The Mann-Whitney Sum Rank is commonly considered one of the strongest nonparametric tests available (Zar 1984). This nonparametric test has been suggested to be 95% as powerful as its parametric counterparts, with power increasing when assumptions of the parametric tests are seriously violated (Zar 1984). Although this test may provide an accurate analysis of these particular data, the inherent weaknesses of nonparametric statistics must be considered when reviewing this study.
Tides
Tidal patterns off Cape San Blas were recorded using 2 Hydrolab DataSonde 3 data loggers that were placed 50–75 m offshore of east and west beaches. A pressure sensor within the data loggers measured tidal height every 15 minutes to the nearest one-hundredth of a meter.
Winds
Wind patterns along Cape San Blas were assessed using data gathered by a National Weather Service C-Man station located on Cape San Blas (National Data Buoy Center, Stennis Space Center, Mississippi). For analysis, wind directions were divided into 8 categories of 45° each: north, northeast, east, southeast, south, southwest, west, and northwest.
Currents
During the 2000 summer season, buoys were deployed weekly at 4 Florida Fish and Wildlife Conservation Commission (FWCC) R-monuments to determine nearshore current patterns and velocities. Two monuments were located on East beach (R-123 and R-121) separated by 0.97 km, and 2 benchmarks were located on West beach (R-110 and R-107) separated by 0.65 km. Buoys consisted of frozen grapefruit which were launched from the water's edge approximately 100 m into the Gulf of Mexico, using a modified slingshot attached to the rear of a 4-wheel drive pickup truck. The buoys were observed as long as possible by personnel onshore. Every 15 minutes, time, distance traveled, and wind speed and direction were recorded. Distance traveled was approximated by measuring the straight-line distance onshore from one observation to the next.
To further estimate direction of sand transported by the longshore current (longshore drift), daily oceanographic observations following those of Schneider and Weggel (1982) were conducted at one FWCC benchmark on East beach and one along West beach from April through August 2000. Data collected included wave period, direction, and type; breaker height; wind speed; ocean current speed and direction; foreshore slope; and width of the surf zone. These data were then used to calculate longshore drift using the equation of Walton (1980), which incorporates fluid density, acceleration of gravity, breaking wave height, width of surf zone, mean longshore current velocity, distance of buoy used to determine current velocity from shore, and a friction factor (0.1). Current direction was divided into 2 categories: north or south. For analysis, current direction was labeled either 1 for north or −1 for south. The relationship between current direction and wind direction was assessed using logistic regression in Minitab (Minitab Inc. 1996). The relationship between current direction and sea turtle nesting was examined by using a Mann-Whitney Rank Sum test.
Topography
Topographical measurements were taken along the West and East beaches of Cape San Blas biweekly during summer (15 May to 1 September) and monthly throughout the remainder of the year. Transects originated at the same 4 FWCC monuments that served as buoy launch sites. Along East beach, the transect at monument R-123 ran at 154° for approximately 110 m and the one at R-121 ran at 143° for approximately 135 m. Along West beach, the transect at monument R-110 ran at 234° for approximately 135 m, and the one at R-107 ran at 220° for approximately 55 m. Heights of the beach were recorded using a laser transit and were documented every 5 m along the transect, as far into the Gulf of Mexico as possible (∼ 20 m). The relationship between sand movement and wind direction was assessed using logistic regression in Minitab (Minitab Inc. 1996). A linear trendline was fit to the shallowest and steepest profile for each year and each benchmark in Microsoft Excel to estimate slope. The mean of the 2 slopes was calculated for an overall slope for each benchmark. The mean slopes of the 2 western benchmarks were averaged to generate an overall mean for West beach, and of the 2 eastern benchmarks for East beach. Comparing slopes of the beach at various times allowed an assessment of erosion/accretion patterns throughout this study.
RESULTS
Sea Turtle Surveys and Reproductive Success
A mean of 65 sea turtle nests were deposited on Cape San Blas in 1998, 1999, and 2000, and of those, a mean of 78.1% was observed at oviposition (Table 1). Of the 111 turtles that were tagged, 27 (24.3%) nested more than once; 8 loggerheads (7.2%) nested 3 or more times. One turtle tagged on Cape San Blas on 15 June 1998 was observed nesting on the eastern end of Gulf Islands National Seashore on Perdido Key, Florida, on 17 July 1998 (Mark Nicholas, Gulf Islands National Seashore, pers. comm. July 19, 1998). These nests were oviposited 32 days apart, with an internesting distance of approximately 250 km. Of the 153 nesting events recorded during this study, 94 (61.4%) nests were deposited on West beach and 59 (38.6%) were deposited on East beach. Along West beach, turtles nested almost equally during east (46.8%) and west (53.1%) winds. On East beach, however, turtles nested more frequently during west winds (80.7%) than east winds (19.3%; t = 187.5, p = 0.004; Fig. 2).
Citation: Chelonian Conservation and Biology 6, 2; 10.2744/1071-8443(2007)6[206:RONSTT]2.0.CO;2
Of all nests deposited on Cape San Blas, hatchling emergence success was 33.5% in 1998, 54.1% in 1999, and 41.5% in 2000 (Table 2). There was significantly greater success in 1999 than in 1998 (t = 3147.5, p = 0.003). In 1998, 55.4% of nests were lost to erosion before the completion of incubation. This percentage declined to 16.3% in 1999 and 30.6% in 2000. One nest was lost to raccoon depredation in each year. Nests that were relocated due to inundation (were not included in hatchling emergence success calculations—refer to Methods) and included 15 (26%) nests in 1998, 10 (13%) nests in 1999, and 15 (24%) nests in 2000.
In 1998, success was greater along East beach (39.8%) than West beach (23.3). In 1999 and 2000, however, success was greater along West beach (56.8%, 1999; 36.5%, 2000) than along East beach (36.9%, 1999; 22.3%, 2000). None of these comparisons was statistically significant (p > 0.01), however, due to small sample sizes and high variability.
Tides
Tidal information was gathered off West beach for 54 days in 1998 and 9 days in 1999, and off East beach for 5 days in 2000. Tidal patterns collected from water monitors off both beaches were nearly identical to those provided by the National Oceanographic and Atmospheric Administration. The diurnal tidal pattern observed off Cape San Blas was synchronous between West and East beaches. Comparison of tidal patterns and timing of sea turtle nesting for all 3 years revealed 98% (152) of turtles nested on a rising tide and 2% (3) on a falling tide. No turtles nested on a falling tide in 1998, 1 turtle did so in 1999, and 2 turtles nested while the tide was falling in 2000.
Winds
Wind direction was recorded every day from May 1998 through August 2000. During the fall and winter, the wind blew primarily from the north and east (north, northeast, east), whereas during the spring and summer it blew mainly from the south and west (south, southwest, west).
Currents
Current speed and direction was observed on 13 days from April 2000 through August 2000. Along East and West beaches, there was a positive relationship between wind and current direction (East z = −3.76, p < 0.001; West z = −2.34, p = 0.019). Results of oceanographic observations also demonstrated this relationship. Observations were collected for 57 days from April through August 2000. Along West beach, the current traveled west on 21 (36.8%) days and east on 36 (63.2%) days. When the current flow was west, the wind blew primarily from the northeast, east, southeast, or south (85.7%), and when it traveled east the wind blew most often from the southwest, west, northwest, or north (81.0%). Along East beach, the current traveled west on 14 (25.4%) days and east on 41 (74.6%) days. When the current flow was westerly, the wind blew from the north, northeast, east, or southeast as often (50.0%) as when it blew from the northwest, west, southwest, or south (50.0%). However, when the current traveled east, the wind blew primarily from the northwest, west, southwest, or south (80.5%).
Topography
From September 1998 to August 2000, West beach lost 4.95 m of sand along the entire profile. Individual points along the profile differed; the greatest loss (−1.17 m) occurred 30 m from the benchmark, whereas the first 15 m of the profile gained 0.16 m. During this period, East beach gained 3.78 m of sand along the entire profile. The greatest gain (0.61 m) occurred 35 m from the benchmark, whereas the greatest loss (−0.18 m) was documented 45 m from the benchmark. The mean slope of West beach was −0.135, whereas that along East beach was −0.060. There was a significant relationship between wind direction and sand movement along East and West beach. On both beaches, accretion occurred more often when the wind blew from the east than when it blew from the west, whereas erosion occurred more often during winds from the west than those from the east (East z = −2.03, p = 0.042; West z = −2.50, p = 0.012). There was no seasonal variation in sand movement.
DISCUSSION
Environmental cues, such as slope, moisture, temperature, and salinity aid in loggerhead turtle nest site selection; however, along dynamic coasts these characteristics are constantly changing. In these areas, offshore characteristics, such as water depth, tides, and currents may help reduce energy expenditure of nesting females and increase reproductive success. Leatherback turtle (Dermochelys coriacea) nesting colonies are associated with beaches that provide deep nearshore access, which may enable turtles to attain high nesting ground with minimal effort (Hendrickson 1980; Eckert 1987). Along Cape San Blas, turtles nested more often along the narrow, eroding West beach than the wide, accreting East beach. This may be due in part to deeper waters off East beach, which helped create a steep beach profile and enabled a nesting turtle to expend less energy getting to an appropriate nesting site.
It seems as if any benefit a turtle would derive from nesting along a narrow, eroding beach would be offset by the lower reproductive rate that eroding beach would most likely provide. However, in this study there was no difference in hatchling emergence success between the accreting East beach and the eroding West beach. Although this area experiences a large number of tropical storms, these events represent only a few days during the entire nesting season. During those days, nests on West beach may be threatened by inundation and erosion. However, under the normal weather conditions that represent the majority of the season, the steep slope of West beach may serve as enough of a barrier against high tides to allow nests to fully incubate. Therefore taking advantage of the deeper waters and steep beach profile of West beach may allow turtles to expend less energy while still producing successful nests.
In addition to nesting on a steep beach profile, nesting when tides are high may decrease the distance required to reach an appropriate nesting site, which may reduce energy expenditure and shorten the time a turtle is exposed to predators (Bustard 1979; Frazer 1983a). It has been suggested that turtles only use this strategy in areas where tidal ranges are great (Frazer 1983a); however, along Cape San Blas, tidal range is less than 0.3 m and turtle nesting is strongly correlated with tidal height. These results may reflect the influence of wind on ocean currents in this area. Winds often blow waters farther up the beach during rising tides, which may increase actual tidal amplitude but not be reflected in tidal data recordings (Stauble and Warnke 1974). In areas where waters are wind-driven, turtles may take advantage of the extra energy provided by winds to rising tides.
Currents may also be used to reduce energy expenditure by nesting turtles. In Japan, loggerhead turtles are often located within the Kuroshio current during the internesting period, which may allow turtles to drift passively and conserve energy for their next nesting attempt (Naito et al. 1990; Sakamoto et al. 1993). Green turtles (Chelonia mydas) off Ascension Island remained relatively stationary after entering the sea following nesting and may rest or drift within local currents during this time (Mortimer and Portier 1989). Conserving energy during internesting periods may allow for increased hatchling emergence success by allowing adult females to reach the most appropriate nesting sites. Along Cape San Blas, turtles may frequent deeper waters off West beach during internesting periods. When winds blow from the west and generate easterly flowing currents, turtles may be carried over the Cape San Blas shoals and into the vicinity of East beach. However, when easterly winds create westward flowing currents, energy must be expended for turtles to swim against the current and over the shoals to nest along East beach. In this dynamic environment, preserving energy and nesting along a steeply sloping beach may help increase reproductive success of loggerhead turtles.
Turtles nesting along unstable beaches may scatter their nests on a water-to-dune axis to maximize reproductive success (Mrosovsky 1983; Eckert 1987; Bjorndal and Bolten 1992; Wood and Bjorndal 2000). Turtles nesting in these regions may also scatter nests laterally along the beach. Eckert (1987) suggested nest dispersal should occur whenever nest survival is not strongly correlated with available environmental information. Typically, loggerhead turtles have lower site fidelity than those species that nest along more stable beaches, such as green and hawksbill (Eretmochelys imbricata) turtles (Carr and Carr 1972; Talbert et al. 1980; Williams-Wallis et al. 1983; Garundo-Andrade 1999). Perhaps this reduction in site fidelity permits turtles nesting along unpredictable coasts to scatter nests throughout the entire nesting beach and reduce losses to erosion and inundation. This strategy may also permit energy conservation by allowing turtles to drift with the current to a nesting location rather than expend energy to swim back to a specific nesting site. Along Cape San Blas, if a turtle originally nested along east beach during an easterly wind and attempted to renest during a westerly wind, it would have to swim against the current to renest in that location. Size of the study site during this research was too small to accurately assess site fidelity of loggerhead turtles nesting along Cape San Blas however. Increasing the sampling area to include the entire St. Joseph Peninsula may help determine whether loggerhead turtles nesting in this dynamic area utilize this strategy to reduce nest loss to erosion and inundation and maximize reproductive success.
Turtles nest along dynamic barrier islands throughout the world and although these areas appear less optimal than more stable beaches, the narrow, highly eroding beaches may help reduce energy expenditure while still producing successful nests. Animals have evolved behavioral patterns that allow them to thrive in many harsh environments, such as dry hot deserts and frozen arctic tundra. Similarly, loggerhead turtles have evolved a pattern that uses cues such as tidal heights and water depth to allow success on extremely dynamic barrier beaches. This strategy may also include reduction of site fidelity, and further research along Cape San Blas and similar sites may support this theory.
Cape San Blas, located on the southern tip of the St Joseph Peninsula in the Florida panhandle, is part of a dynamic barrier island system that supports nesting sea turtles. This barrier island system extends along the northern Gulf of Mexico and although these beaches experience extreme erosion and accretion, they support a significant group of nesting loggerhead turtles.
The relationship between wind direction, current direction, and loggerhead turtle nesting along Cape San Blas, Florida, from June 1998 through August 2000, as shown on an aerial photograph of Cape San Blas. Winds blowing from the east (a) resulted in a westward flowing current. Under these conditions, significantly fewer turtles nested along East beach than when winds blew from the west (b). Winds from the west caused easterly flowing currents and resulted in a larger number of nests deposited along East beach. Aerial photograph courtesy of US Geological Survey.
