Declining Reproductive Parameters Highlight Conservation Needs of Loggerhead Turtles (Caretta caretta) in the Northern Gulf of Mexico
Abstract
Marine turtles in the Gulf of Mexico are at risk due to many anthropogenic threats including habitat degradation, commercial fishing, and petroleum activities. The severity of this risk was made apparent in 2010 with the occurrence of the Deepwater Horizon oil spill. The objectives of this study were to assess long-term trends in abundance and reproductive parameters for this genetically distinct nesting group. From 1994 to 2010, morning surveys were conducted along 3 beaches on the St Joseph Peninsula, Florida, including within our primary study site on Cape San Blas. Nest abundance on all 3 beaches declined by at least 47% (p < 0.01). Mean nesting success on Cape San Blas was 40% and also declined (p = 0.002). Mean clutch size was 108 and mean emergence success was 58%. Throughout the study there were no changes in clutch size and emergence success. We found that nesting characteristics for the northern Gulf of Mexico subpopulation appear similar to those from other loggerhead turtle nesting groups in the southeastern United States in some ways, such as emergence success, timing of peak nesting, and incubation duration and different in other ways such as nesting success. Variation in some of the parameters may indicate turtles among the different nesting groups experience different environmental conditions. The severity of declines in nest abundance and the low nesting success reported for this small subpopulation suggest potentially serious consequences for this nesting group.
Loggerhead turtles (Caretta caretta) that nest along the Gulf of Mexico coast face many threats to survival including commercial fishing (NMFS 2008), pollution (Alam and Brim 2000), boat-related mortalities (Sobin 2008), and loss of nesting habitat (Lamont and Carthy 2007; Hawkes et al. 2009; Witherington et al. 2009). Marine animals in this region face the additional and serious threat of the largest oil spill in US history. Because of these threats, Bjorndal et al. (2011) suggest critical data, including demographic and abundance parameters, must be collected to assess and limit long-term declines in populations. In addition, the most recent report by the Turtle Expert Working Group (TEWG) highlights as a top priority the identification of population parameters for all loggerhead nesting groups (TEWG 2009).
One of the largest nesting aggregations of loggerhead turtles in the Atlantic basin is found along the southeastern United States where about 80% of all nesting occurs and 90% of all hatchlings are produced (TEWG 2009). The large contribution of this region to the global population highlights the importance of understanding the status of this group. Genetic studies have divided the Western Atlantic loggerhead nesting group into 5 subpopulations: 1) northern (Florida/Georgia border to southern Virginia), 2) peninsular Florida (Florida/Georgia border through Pinellas County, Florida), 3) Dry Tortugas (islands west of Key West, Florida), 4) northern Gulf of Mexico (Franklin County, Florida, through Texas), and 5) greater Caribbean (all other nesting beaches throughout the Caribbean and Mexico) (Encalada et al. 1998; TEWG 2009; Shamblin et al. 2011). Studies have indicated that the abundance of loggerhead nests along the Atlantic coast and in southwestern Florida is declining (Dodd and Mackinnon 2008; TEWG 2009; Witherington et al. 2009). Continued declines in these numbers was a primary factor prompting a proposal for elevating the level of protection for loggerheads in the United States (USFWS and NMFS 2010) and in the recent designation of 9 distinct population segments of loggerheads as either Endangered or Threatened (Wallace et al. 2010; USFWS and NMFS 2011), thereby highlighting the importance of developing population assessments for each of these subgroups.
Assessments of abundance are difficult for long-lived and wide-ranging species such as marine turtles (Frazer 1983; Crouse et al. 1987; Heppell et al. 2000, 2003; Witherington et al. 2009). The life cycle for the loggerhead turtle requires multiple habitats and spans ocean basins: eggs are deposited on beaches, hatchlings remain in oceanic currents, juveniles use coastal bays, and adults migrate across oceans (Dodd 1988). Therefore, population assessments for marine turtles are typically generated using data from the most accessible life-history stage, nesting females, using parameters such as nest abundance, clutch size, and emergence success (Dodd and McKinnon 2008; Witherington et al. 2009; Bolten et al. 2011). Unfortunately, these data have previously been lacking for turtles nesting in the northern Gulf of Mexico, thereby making population assessments for this nesting group difficult (Wallace et al. 2010). Our goals were to examine temporal trends in nesting and hatching parameters for loggerhead turtles in the northern Gulf of Mexico, including abundance, nest success, clutch size, and emergence success over a 17-yr period.
METHODS
Study Site
We conducted intensive nesting surveys and calculated emergence success from 1994 to 2010 along 5 km of beach on Eglin Air Force Base property along Cape San Blas (CSB), Florida, which represents the southern tip of the St Joseph Peninsula (SJP). This area supports one of the greatest nesting densities of loggerhead turtles in the northern Gulf of Mexico (Lamont and Carthy 2007). Nest abundance data from adjacent beaches on 7 km of the SJP and along 16 km of beach within T.H. Stone Memorial St Joseph State Park (SJSP; data made available by T. Spector, Florida State Parks; Fig. 1) were also used to calculate trends.
Citation: Chelonian Conservation and Biology 11, 2; 10.2744/CCB-1006.1
Nesting and Hatching Surveys
Surveys were conducted every morning during the nesting season from 1 May to 1 November on foot, all-terrain vehicle, or 4-wheel drive vehicle. All turtle crawls were identified to species and egg clutches were located. Four wooden stakes were placed around the clutch, and flagging was wrapped around the stakes. On CSB, intensive morning surveys began in 1994 and data collected included number of crawls, number of nests, clutch size, incubation period, and emergence success. On SJP, annual nest abundance data were available from 1995, whereas in SJSP data were available beginning in 1996. All surveys followed standard protocol established by the Florida Fish and Wildlife Conservation Commission (FWC 2007; http://www.myfwc.com/media/418106/Seaturtle_Guidelines.pdf).
We defined nesting success as a proportion of the total number of nests to the total number of crawls. On CSB, nests were excavated 3 d posthatching or after 75 d of incubation. The total number of eggs in the clutch was assessed during nest relocation. Clutches were relocated landward of the original nest site if they were immediately threatened by erosion or long-term inundation. Incubation length was defined by calculating the number of days between deposition (determined during morning survey) and the date of first emergence (determined by presence of hatchling tracks). If rain or high winds could have obscured those first signs of emergence, nests were not included in analyses. 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. Emergence success was defined as the number of hatchlings that emerged from the nested divided by the total number of eggs deposited in the nest, following Johnson et al. (1996).
Analyses
We first examined the temporal trends in annual nest abundance along all 3 beaches using a Poisson regression. For CSB, we also examined variation in nest abundance by month (May–August) and temporal trends in monthly nest abundance and clutch size using Poisson regression.
Nesting success was modeled using year and month as independent variables. Emergence success was modeled with year and month effects. Both nest success and emergence success were analyzed using logistic regression. We used the GENMOD procedure of SAS 9.2 for Poisson and logistic regressions and used an α-level of 0.05 to assess significance of covariate effects.
RESULTS
At all 3 sites on the SJP, the number of nests significantly declined over time; based on the fitted model, the estimated decline was 46.8% (1994–2010), 55.8% (1996–2010), and 46.8% (1995–2010) declines in CSB (Nnest = 4.106 − 0.039 × year, pslope < 0.01), SJP (Nnest = 5.359 − 0.055 × year, pslope < 0.01), and SJSP (Nnest = 5.073 − 0.045 × year, pslope < 0.01; Fig 2).
Citation: Chelonian Conservation and Biology 11, 2; 10.2744/CCB-1006.1
The mean number of crawls that occurred along CSB was 109, with 43 (8.7 nests/km) of those crawls resulting in a nest and 63 (12.6/km) resulting in no nest (Table 1). During the entire study period only 1 nest was deposited in September, therefore this data point was combined with August. The mean number of nests laid in May (3) and August (2) was smaller than the mean number laid in July (20; p < 0.01; Table 2); however the mean number deposited in June (19) was similar to the mean number laid in July.
Mean nesting success was 40.1% with greatest success (56%) occurring in 1999 and lowest success (23%) occurring in 2006. Nesting success did not differ within the season (p > 0.05; Table 1) but there was a significant declining trend in nesting success throughout the study (p < 0.01; Table 1).
Loggerheads deposited a mean of 108 eggs in each nest (Table 1). There was no change in clutch size throughout the study (p = 0.30; Table 2). However, clutch size of nests deposited in May tended to be smaller (p < 0.01; Table 2) and those deposited in August tended to be larger (p < 0.01; Table 2) than those deposited in July.
Mean emergence success was 58% (Table 1). There was no significant trend in emergence success during the study period (p = 0.56; Table 2); however, there were monthly differences with greater emergence success in May and June than in July and lower emergence success in August than in July (p < 0.01; Table 2). Eggs incubated for an average 61 d before hatching (Table 1).
DISCUSSION
Results of this study support those of Witherington et al. (2009) who suggest a decline in loggerhead nesting along the Atlantic coast and in southwestern Florida; however the dramatic decline in nest abundance along the SJP and in nesting success along CSB suggest this small subpopulation in the northern Gulf of Mexico may also be at risk. Estimated declines in nest abundance on the Atlantic and southwestern coasts of Florida ranged from 29% to 37% between 1989 and 2006 (Witherington et al. 2009), whereas abundance of nests along the northern Gulf of Mexico declined by almost half from 1994 to 2010. This nesting group is one of the smallest subgroups within the larger southeastern US nesting population and the number of females contributing to this subpopulation is unknown. It is believed that extinction of a species occurs when mortality rates exceed reproductive replacement over a sustained period (Sodhi et al. 2009). If this large decline in nest abundance along the northern Gulf of Mexico is an indication of loggerhead mortality and if nesting success is used to define reproductive replacement, then there may be serious consequences for this nesting group. Findings from our study highlight the urgent conservation needs for this nesting group.
We also found that nesting characteristics for the northern Gulf of Mexico subpopulation appear similar to those of other loggerhead turtle nesting groups in the southeastern United States in many ways, such as emergence success, timing of peak nesting each season, incubation duration, and clutch size, but different in nesting success (Table 3). During our study, nesting occurred most often in June and July, which coincides with peak nesting throughout the southeastern United States (Weishampel et al. 2004; Antworth et al. 2006). Emergence success within our study site also appears to be similar to that reported on other nesting beaches in the southeastern United States including in Brevard County, Florida, and along the southwest Florida coast (Ehrhart et al. 2003; Tucker et al. 2009). Emergence success is often affected by sand characteristics (Fadini et al. 2011) such as hardness (Kudo et al. 2001) and mean particle diameter (Mortimer 1990). Clutch size fell within the range reported for other studies (Hedges 2007; EAI 2008). Because clutch size has been linked to turtle size, this suggests turtles nesting in the northern Gulf are of similar size to those nesting throughout the southeastern United States (Frazer and Richardson 1985; Hays and Speakman 1993; Broderick and Godley 1996). Our study site is a natural beach that has never been nourished, therefore sand characteristics may be optimal for successful hatchling incubation and emergence. Finally, because incubation duration is determined by incubation temperatures, incubation duration at our study site appeared more similar to durations reported for cooler, northern nesting beaches such as those in North Carolina (Rush 2003) and Georgia (Dodd and Mackinnon 2008). Incubation durations on warmer beaches, such as those in southwestern Florida (Tucker et al. 2009) and the Everglades, Florida (Davis and Whiting 1977), appear to be longer.
Although some aspects of loggerhead reproduction appear similar between the northern Gulf of Mexico and other nesting groups in the southeastern United States population, nesting success appeared to be lower than that reported for many other beaches in the southeastern United States, including those in Georgia and Brevard County, Florida (Hawkes et al. 2005; Antworth et al. 2006; Brock et al. 2009; Tucker et al. 2009; Witherington et al. 2009). Low nesting success indicates increased false crawling behavior. False crawling behavior has been linked to activities such as turtle walks (Johnson et al. 1996), artificial lighting (Stoneburner and Richardson 1981), and changes in beach substrate due to nourishment (Raymond 1984). However because the beach along CSB is uninhabited and has never been nourished, false crawling behavior at our study site may not be attributed to direct anthropogenic causes. Turtles attempting to nest may also be disturbed by predators, such as coyotes and raccoons, thereby causing them to false-crawl (Shoop et al. 1985). From 1994 to 1997, coyote depredation of turtle nests on CSB was fairly common and resulted in destruction of approximately 20%–40% of nests each year. However in 1998, consistent predator control activities began on Eglin Air Force Base and within nearby SJSP, which has since reduced predation of turtle nests at our study site to a handful (∼ 5–10) of eggs taken each year by ghost crabs (Ocypode quadrata), fish crows (Corvus ossifragus), and fire ants (Solenopsis invicta; Parris et al. 2002). Because nesting success did not improve following initiation of predator control, it indicates that predators are not causing turtles to false-crawl on CSB.
It has been suggested that turtles use cues located both offshore and onshore to identify nesting sites. Offshore factors include the presence of reefs and rocks (Hughes 1974), magnetic fields (Lohmann and Lohmann 1996), and olfactory cues (Carr 1972), whereas onshore cues such as temperature, salinity, slope, moisture, width, and sand type have been suggested to influence nest site selection (Johannes and Rimmer 1984; Provancha and Ehrhart 1987; Kikukawa et al. 1996; Garmestani et al. 2000; Wood and Bjorndal 2000). Our study site on CSB is located on a highly dynamic cape spit. This area experiences one of the greatest rates of natural erosion in Florida (Lamont and Carthy 2007). Perhaps the dynamic nature of the nearshore environment off CSB blurs cues used in nest site selection. Lack of offshore cues may result in a greater number of false crawls if turtles must come onshore to assess site selection.
Quantitative demographic analyses require large amounts of data on life-stage–specific parameters such as survival, growth and fertility, migration, and on the effects of environmental variability on these parameters (Groom and Pascual 1998; Heppel et al. 2003). Loggerhead turtles nesting in the northern Gulf of Mexico experience different environmental variables than those that nest along the Atlantic coast, therefore the strategies they use to optimize reproduction most likely differ. Results from our study demonstrate that parameters among nesting groups within one population can vary greatly (Wallace et al. 2010), which highlights the importance of gathering life-history parameters for development of accurate population models and effective conservation strategies. This is particularly important for this group of loggerhead turtles that, due to large declines in nest abundance and decreasing nest success, requires immediate management and conservation action.
Three nesting beaches on the St Joseph Peninsula, Florida, located in the northern Gulf of Mexico, where reproductive parameters for loggerhead turtles were collected from 1994 to 2010.
The number of nests laid per year on adjacent beaches along the St Joseph Peninsula (SJP), Florida, in the northern Gulf of Mexico: Eglin Air Force Base (EAFB) property on Cape San Blas from 1994 to 2010, the SJP from 1995–2010, and within St. Joseph State Park (SJSP) from 1996 to 2010.
