seaturtle.org : MTN : ARCHIVES : Sign In

Here, we evaluated satellite-tracking data for a female hawksbill tracked from nesting to foraging grounds and back that showed flexibility in nesting beach selection at a local level. Specifically, we aimed to 1) delineate home range and core-use areas for both inter-nesting and foraging grounds and compare them across years, 2) determine if migration routes showed repeatability and 3) assess variations in nesting beach selection.

In the Caribbean, hawksbills tend to nest in low densities and there are few beaches that support high densities (>500 crawls per year) of hawksbill nesting (Dow Piniak & Eckert 2001). Buck Island Reef National Monument (BIRNM) is one such high-density nesting beach and is the only fully protected area where hawksbills both forage and nest (USFWS & NMFS 1993). BIRNM includes a 0.71 km2 uninhabited island (Buck Island) located on the shallow St. Croix shelf (approx. 15-20 m depth range), 2.4 km northeast of the island of St. Croix in the US Virgin Islands (USVI) (Fig 1; 17.78° N, 64.62° W). The National Park Service (NPS) has conducted a saturation-tagging sea turtle nesting and monitoring program at BIRNM for over 25 years (Z. Hillis-Starr, unpubl. data).

As part of a larger, ongoing satellite-tagging effort to study inter-nesting periods, migrations, and foraging areas of nesting hawksbills (K.M. Hart, unpubl. data; also see <https://seaturtle.org/tracking/?project_id=663>), we satellite-tagged a nesting female hawksbill (Turtle TTZ531) at BIRNM on 7 August 2012, following established protocols (NMFS SEFSC 2008). Turtle TTZ531 was captured on the south shore of Buck Island (Fig. 1) and measured 84.9 cm straight-carapace length (nuchal notch to most posterior marginal scute tip). This turtle had not previously been intercepted nesting at Buck Island, but an existing flipper tag confirmed she had nested on a beach in Jack’s Bay, St. Croix in September 2007, approximately 5 years earlier and 10 km distant (P. Eliazar pers. comm.; see Fig. 1). We chose to highlight this turtle because she stood out from the other tagged hawksbills in two ways: 1) she was our longest satellite-tracked turtle (852 days as of data synthesis) and 2) she was one of only two hawksbills that we tracked back to BIRNM after migrating away to distant foraging grounds. Both turtles returned to the same local area of the USVI during the nesting season, but unlike the other turtle, Turtle TTZ531 did not nest at Buck Island upon her return.

We intercepted Turtle TTZ531 after successful nesting and fitted a Wildlife Computers SPOT5 platform terminal transmitter (PTT) to the carapace (2.5xAA model, 71 mm long x 54 mm wide x 24 mm high). Prior to transmitter application with SuperBond™ two-part cool-setting marine epoxy, we removed epibionts (e.g., barnacles, algae), then sanded and cleaned the carapace with isopropyl alcohol. We streamlined attachment materials and minimized the epoxy footprint. The tag had a saltwater switch and output of 0.5 W and a mass of 115 g in air. The anticipated battery life was 1 year and it was programmed to operate every day from 1 June to 30 November and every third day from 1 December to 30 May. The turtle was released within 2 hours at the capture location.

We filtered satellite data through the Satellite Tracking and Analysis Tool (STAT, Coyne & Godley 2005) on <https://seaturtle.org>. Argos assigns accuracy estimates of <250 m for location class (LC) 3, 250 to <500m for LC2, 500 to <1500m for LC1, and >1500m for LC 0 (CLS America 2011). The estimated accuracy is unknown for LCs A and B, and locations failing the Argos plausibility tests are tagged as class LC Z. Argos performed Kalman- filtering (initiated in 2011, Kalman 1960) on location data. This newly-implemented Kalman- filtering algorithm provides more estimated positions and significantly improves position accuracy, most significantly for locations obtained in LCs A and B (Lopez & Malardé 2011). Data were analyzed using a switching state-space model (SSM), which predicts a behavioral mode and location every 8 hours. The predicted mode was binary and categorized as either foraging or migration (e.g., Jonsen et al. 2005, 2007). However, because we tagged the turtle during the nesting season, behavioral modes outside of migration were defined as either inter-nesting or foraging based on point location and time of year. With the time frames for inter-nesting and foraging identified by SSM, we used raw satellite location data received during these times for further analysis.

We ran the SSM, filtered satellite data, conducted site fidelity tests and created kernel density estimates (KDE) in the same way as for our previous studies (see Hart et al. 2013, 2014). We filtered out locations based on speed (>5 km/h), LC Z, topography (points on land) and very distant locations (> 120 km from nearest valid point). For foraging periods, we also filtered out locations deeper than -200 m bathymetry (neritic zone), which accounted for less than 0.01% of all filtered locations. We did not filter locations by depth from inter-nesting periods as the area around BIRNM consists of rapidly changing bathymetry. We used the GEBCO_08 Grid (General Bathymetric Chart of the Oceans), a 30 arc-second continuous terrain model of both ocean and land to map bathymetry contours. For site fidelity tests, we bounded the random walks by land and bathymetry (foraging areas: -200 m, inter-nesting areas: -4,500 m) to create boundaries that included all filtered locations. We generated mean daily locations from filtered data during non-migration times for KDE analysis. For inter-nesting and foraging periods that passed site fidelity and had at least 20 mean daily locations, we used 95% KDEs to represent the home range and the 50% KDEs to represent the core area of activity. We generated centroids for 50% KDEs, and we used the largest activity center if a 50% KDE had multiple centers. We summarized the distance between centroids as well as the depth and distance to the nearest shoreline. For migration periods, we summarized time in migration as well as plotted and measured the paths for both straight-line and cumulative distances.

The SSM predicted both migration and non-migration (inter- nesting or foraging) modes throughout the turtle’s tracking period. Turtle TTZ531 spent 69 days at BIRNM after tagging and then migrated for 30 days to the Bahamas where she spent 561 days. She then migrated back to BIRNM and spent 82 days there during inter- nesting. After this, she once again migrated back to the Bahamas (Table 1).

Table 1. Dates and switching state-space model (SSM) behavioral modes during entire tracking period for Turtle TTZ531, an adult female hawksbill satellite-tagged while nesting within BIRNM, USVI. KDE = kernel density estimation, IN = inter-nesting, MIG = migration, F = foraging. The number of days may include small variations in mode, such as short (1 day) foraging stop-overs during migration or any short migrations (1-4 days) at inter-nesting or foraging areas. No KDE analysis was done with migrating data locations. ^aDue to a short mode switch (4 days in migration mode) during the first foraging time at the Bahamas, the KDE for this time is from 5 December 2012 to 28 May 2014 only. ^bTransmissions continued from same general location after data synthesis date (6 December 2014) until transmission ceased on 25 April 2015.

Figure 1. Buck Island Reef National Monument (BIRNM; black line) with yellow star denoting Buck Island, the original satellite-tagging location. Turtle TTZ531 also nested in 2007 at Jack’s Bay (black star). Inter-nesting (IN) kernel density estimates (KDEs; green and red polygons) shown with centroid locations (matching colored points). Bottom right inset includes high-quality locations (LCs 1, 2, 3) during IN 2; yellow points show likely nesting locations (19 August, 4 September and 20 September 2014; area shown in main panel by hollow star) away from the main nesting beach of Buck Island.

We tracked Turtle TTZ531 through the initial inter-nesting time (IN 1) during 2012 and then again through a second inter-nesting period (IN 2) at BIRNM beginning approximately two years later in July 2014. During her second presumed inter-nesting time (nesting was not intercepted), she was not encountered on Buck Island by the saturation tagging and monitoring program. High-quality tracking locations on land approximately 2 weeks apart (LCs 1, 2 and 3; see Fig. 1) during this time however, indicated a high probability of nesting on the north coast of St. Croix (near the Estates of Judith’s Fancy beaches where sea turtle monitoring and tagging does not take place), approximately 12 km away from Buck Island (Fig. 1). Despite the separation of these beaches, the inter-nesting core-use and home range areas overlapped substantially (Fig. 1) and were of similar size (Table 1). IN1 and IN2 core-use areas had 16.5 km2 of spatial overlap, which constituted 85% of IN1 and 99% of IN2 and the 50% KDE centroids were only 0.3 km apart. Inter-nesting centroids were both located in areas with shallow bathymetry (-3 m) and were similarly close to shore (< 0.6 km).

We tracked Turtle TTZ531’s migration (MIG) between the USVI and the Bahamas on three separate occasions. Travel times ranged from 30-36 days (Table 1 and Fig. 2). The straight-line distance between her inter-nesting and foraging area centroids averaged 1,355.5 km (± SD = 7.4 km, n = 3), but the cumulative distance between each filtered location along her path was closer to 2,000 km (mean ± SD = 1,929.4 ± 83.1 km, n = 3).

Figure 2. The entire track of filtered locations received from Turtle TTZ531, an adult female hawksbill satellite-tagged while nesting at BIRNM, US Virgin Islands (USVI). Data points are color-coded by switching state-space model behavioral mode: IN = inter-nesting, F = foraging and MIG = migration. For dates corresponding to each mode, see Table 1.

We tracked Turtle TTZ531 to a foraging area in the Bahamas twice. The turtle remained within the foraging area (F1) for 561 days. She inhabited the second foraging area (F2) for 36 days and continued to transmit from the site at the time of data analysis (6 Dec 2014). These residence/foraging areas were similar in location (Fig. 3) and size (Table 1), with centroids of 50% KDEs measuring only 12.9 km apart. The centroids also had similar bathymetry measurements (F1: -3 m; F2: -2 m) and distances to the nearest shoreline (F1: 85.8 km; F2: 76.9 km).

Figure 3. Foraging (F) kernel density estimates (KDEs; purple and orange polygons) and centroids in the Bahamas for Turtle TTZ531, an adult female hawksbill satellite-tagged while nesting at BIRNM, USVI. For dates corresponding to foraging periods, see Table 1.

We observed Turtle TTZ531 having a somewhat predictable nesting/migration/foraging cycle. Her inter-nesting and foraging areas were markedly similar over the years and she re-migrated to her inter-nesting, in-water habitat on the expected 2–year interval. While little is known about across-season repeatability of in-water residence areas for hawksbills, Caribbean hawksbills tracked through remigration to nesting grounds have been shown to return to the same in-water inter-nesting areas (Dominican Republic; Hawkes et al. 2012). Additionally, multiple females nesting at Buck Island used similar inter-nesting habitat (Starbird et al. 1999) and hawksbills in Barbados used the same inter-nesting residence areas during multiple inter-nesting intervals within the same season (Walcott et al. 2012). Individual fidelity to specific foraging sites has also been demonstrated for hawksbills (Hawkes et al. 2012) and other sea turtle species. For example, individual loggerheads (Caretta caretta) have been tracked after multiple nesting seasons to the same residence areas in the Bahamas (Hart et al. 2015) and in the Mediterranean Sea (Schofield et al. 2010).

Two-year remigration intervals have been reported as the most common for nesting hawksbills on various Caribbean beaches, including Buck Island (Hillis 1994), Antigua (Richardson et al. 1999), and Barbados (Beggs et al. 2007), although much longer remigration intervals have been documented (e.g., 6 years in Barbados (Beggs et al. 2007)). However, as Turtle TTZ531 did not nest again at the monitored beach where she was tagged, her nesting activity upon remigration would have been missed without satellite tracking. If the turtle were to nest within BIRNM on the next 2-year cycle, she would likely be given an incorrect remigration interval of 4 years which would have ramifications for population modeling, as she may have been classified as a non-nester for 2014. Data collected from this turtle’s movements and nesting therefore highlight the variability in nesting behavior among hawksbills, which supports other findings of behavioral plasticity in inter-nesting behavior of Caribbean turtles (Esteban et al. 2015).

One major question concerning sea turtle natal philopatry is how large an area would be considered the “home” beach; this has been shown to vary based on species and geography (see Bowen & Karl 2007). Genetic evidence for hawksbills supports natal philopatry with the possibility of additional nearby nesting sites (see Bowen & Karl 2007). Spreading eggs across different beaches may offer an evolutionary advantage as beach type influences the proportion of hatchlings that survive (Lee & Hays 2004). Other Caribbean hawksbills have shown nesting beach variance: a hawksbill tagged after nesting in the Lesser Antilles traveled 64 km to nest on Anguilla and 187 km to nest at NW St. Croix during the same nesting season (Esteban et al. 2015). Our results support that additional nearby nesting sites may play an important role for hawksbills. Satellite telemetry, in addition to saturation tagging at an index beach, can therefore be used to address fundamental questions on hawksbill nesting and migration behavior. Understanding the frequency that hawksbills use other beaches would be highly valuable for population modeling, demographic analyses and ultimately conservation efforts such as identifying important proximal beaches in need of monitoring and protection.

Despite showing a variable use of nesting beaches, we found Turtle TTZ531 used the same area within BIRNM for both inter-nesting periods across tracking years. BIRNM offers a protected reef with reduced human impacts, such as fishing restrictions and no light or point-source pollution from Buck Island. The waters north of Buck Island also offer access to a deep-water shelf (see bathymetry contours in Fig. 1) which is not available in Buck Island Channel to the south of the island. Further fine-scale habitat assessments could determine if the reefs within BIRNM offer more suitable inter-nesting habitat, such as preferable resting sites. The only other inter-nesting study on hawksbills tagged on Buck Island showed that 7 adult females stayed within 3 km of Buck Island during their inter-nesting periods (Starbird et al. 1999). This further supports that the habitat within BIRNM is important for nesting hawksbills.

BIRNM provides protection to turtles during inter-nesting, however this turtle migrated thousands of miles across Caribbean waters, and could be vulnerable to many threats along the way including direct harvest. Thirteen Caribbean nations allow some form of hawksbill harvest, including ones this turtle passed during migration such as the British Virgin Islands, Haiti and the Turks and Caicos (Dow et al. 2007). Occasional harvesting of nesting females also takes place in Cuba (Dow et al. 2007; see Hart et al. 2012). Direct harvest of hawksbills is not legal in the Bahamas, but little in-water habitat is protected there, and the current fisheries may represent bycatch threats for hawksbills (see Hart et al. 2015). In addition to evaluating migratory corridors and assessing potential threats during different life stages, satellite telemetry offers a tool to track turtles for multiple years and over multiple foraging and inter-nesting cycles, thus providing a supplement to documenting returns at the nesting beach. Satellite-tracking a subset of nesting turtles could highlight exceptions and help determine what proportion of a nesting group may be using other beaches. This in turn can inform population and remigration interval estimates. With concerns such as climate change, nesting beach degradation, threats at sea, and/or low population sizes, understanding these dynamics will be even more important as populations are monitored into the future.

Acknowledgements. We thank J. Beauchamp, M. Denton, A. Daniels, and B. Smith for assistance deploying satellite tags in the field and Buck Island Sea Turtle Research Program assistants in 2014 for help facilitating the program. Permission to tag and sample turtles was given under BUIS permit BUIS-2012-SCI-0002 and USGS-SESC-IACUC 2011-05, issued to KMH. Funding was provided by the USGS Natural Resources Protection Program, Priority Ecosystem Studies Program, and Ecosystems Program. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the US Government.

BASS, A.L., D.A. GOOD, K.A. BJORNDAL, J.I. RICHARDSON, Z.M. HILLIS, J.A. HORROCKS & B.W. BOWEN. 1996. Testing models of female reproductive migratory behaviour and population structure in the Caribbean hawksbill turtle, Eretmochelys imbricata, with mtDNA sequences. Molecular Ecology 5: 321-328.

BOWEN, B.W., W.S. GRANT, Z. HILLIS-STARR, D.J. SHAVER, K.A. BJORNDAL, A.B. BOLTEN & A.L. BASS. 2007. Mixed- stock analysis reveals the migrations of juvenile hawksbill turtle (Eretmochelys imbricata) in the Caribbean Sea. Molecular Ecology 16: 49-60.

BOWEN, B.W. & S.A. KARL. 2007. Population genetics and phylogeography of sea turtles. Molecular Ecology 16: 4886-4907.

BEGGS, J.A., J.A. HORROCKS & B.H. KRUEGER. 2007. Increase in hawksbill sea turtle Eretmochelys imbricata nesting in Barbados, West Indies. Endangered Species Research 3: 159-168.

CLS (COLLECTE LOCALISATION SATELLITES). 2011. Argos user’s manual: worldwide tracking and environmental monitoring by satellite. 19 August 2011 update. <http://www.argos-system.org/web/en/76-user-s-manual.php>. Accessed 9 June 2013.

COYNE, M.S. & B.J. GODLEY. 2005. Satellite tracking and analysis tool (STAT): an integrated system for archiving, analyzing and mapping animal tracking data. Marine Ecology Progress Series 301: 1-7.

DIAMOND, A.W. 1976. Breeding biology and conservation of hawksbill turtles, Eretmochelys imbricata L., on Cousin Island, Seychelles. Biological Conservation 9: 199-215.

DOW, W., K. ECKERT, M. PALMER & P. KRAMER. 2007. An atlas of sea turtle nesting habitat for the Wider Caribbean Region. The Wider Caribbean Sea Turtle Conservation Network and The Nature Conservancy. WIDECAST Technical Report No. 6. <http://seamap.env.duke.edu/seamap2.5/widecast/references/Dow_et_al_2007.pdf>

DOW PINIAK, W.E. & K.L. ECKERT. 2001. Sea turtle nesting habitat in the Wider Caribbean Region. Endangered Species Research 15: 129-141.

ESTEBAN, N., R.P. VAN DAM, E. HARRISON, A. HERRERA & J. BERKEL. 2015. Green and hawksbill turtles in the Lesser Antilles demonstrate behavioural plasticity in inter-nesting behavior and post-nesting migration. Marine Biology 162: 1153- 1163.

HART, K.M., M.M. LAMONT, A.R. SARTAIN & I. FUJISAKI. 2014. Migration, foraging, and residency patterns for Northern Gulf loggerheads: implications of local threats and international movements. PLoS ONE 9: e103453.

HART, K.M., M.M. LAMONT, A.R. SARTAIN, I. FUJISKAI & B.S. STEPHENS. 2013. Movements and habitat-use of loggerhead sea turtles in the northern Gulf of Mexico during the reproductive period. PLoS ONE 8: e66921.

HART, K.M., A.R. SARTAIN & I. FUJISAKI. 2015. Bahamas connection: residence areas selected by breeding female loggerheads tagged in Dry Tortugas National Park, USA. Animal Biotelemetry 3: 1-17.

HART, K.M., A.R. SARTAIN, I. FUJISAKI, H.L. PRATT JR., D. MORLEY & M.W. FEELEY. 2012. Home range, habitat use, and migrations of hawksbill turtles tracked from Dry Tortugas National Park, Florida, USA. Marine Ecology Progress Series 457: 193-207.

HAYS, G.C. 2000. The implications of variable remigration intervals for the assessment of population size in marine turtles. Journal of Theoretical Biology 206: 221-227.

HAWKES, L.A., J. TOMÁS, O. REVUELTA, Y.M. LEON, J.M. BLUMENTHAL, A.C. BRODERICK, M. FISH, J.A. RAGA, M.J. WITT & B.J. GODLEY. 2012. Migratory patterns in hawksbill turtles described by satellite tracking. Marine Ecology Progress Series 461: 223-232.

HILLIS, Z.M. 1994. The first five years at Buck Island Reef Monument - the hawksbill story. In: Schroeder, B.A. & B.E. Witherington (Comps). Proceedings of the 13th Annual Symposium on Sea Turtle Biology and Conservation. NOAA Tech Memo NMFS-SEFSC-341. pp. 242-245.

JONSEN, I.D., J. MILLS-FLEMMING & R.A. MYERS. 2005. Robust state-space modeling of animal movement data. Ecology 86: 2874-2880.

JONSEN, I.D., R.A. MYERS & M.C. JAMES. 2007. Identifying leatherback turtle foraging behaviour from satellite-telemetry using a switching state-space model. Marine Ecology Progress Series 337: 255-264.

KALMAN, R.E. 1960. A new approach to linear filtering and prediction problems. Journal of Basic Engineering 82: 35-45.

LEE, P.L.M. & G.C. HAYS. 2004. Polyandry in a marine turtle: females make the best of a bad job. Proceedings of the National Academy of Sciences 101: 6530-6535.

LOPEZ, R. & J. MALARDE. 2011. Improving ARGOS Doppler location using Kalman ltering. Collecte Localisation Satellites. CLSDT-MEMO-11-65. <http://www.argos-system.org/files/pmedia/public/r282_9_technical_paper.pdf>

MEYLAN, A. & M. DONNELLY. 1999. Status justification for listing the hawksbill turtle (Eretmochelys imbricata) as Critically Endangered on the 1996 IUCN Red List of Threatened Animals. Chelonian Conservation & Biology 3: 200-224.

NMFS SEFSC 2008. Southeast Fisheries Science Center sea turtle research techniques manual. NOAA Tech Memo NMFS- SEFSC-579. <http://www.sefsc.noaa.gov/turtles/TM_579_SEFSC_STRTM.pdf>

NMFS & USFWS. 1993. Recovery plan for hawksbill turtles in the US Caribbean Sea, Atlantic Ocean, and Gulf of Mexico. <http://www.nmfs.noaa.gov/pr/pdfs/recovery/turtle_hawksbill_atlantic.pdf>

RICHARDSON, J.I., R. BELL & T.H. RICHARDSON. 1999. Population ecology and demographic implications drawn from an 11-year study of nesting hawksbill turtles, Eretmochelys imbricata, at Jumby Bay, Long Island, Antigua, West Indies. Chelonian Conservation & Biology 3: 244-250.

SCHOFIELD, G., V.J. HOBSON, S. FOSSETTE, M.K.S. LILLEY, K.A. KATSELIDIS & G.C. HAYS. 2010. Fidelity to foraging sites, consistency of migration routes and habitat modulation of home range by sea turtles. Diversity and Distributions 16: 840-853.

STARBIRD, C.H., Z. HILLIS-STARR, J.T. HARVEY & S.A. ECKERT. 1999. Internesting movements and behavior of hawksbill turtles (Eretmochelys imbricata) around Buck Island Reef National Monument, St. Croix, US Virgin Islands. Chelonian Conservation & Biology 3: 237-243.

VELEZ-ZUAZO, X., W.D. RAMOS, R.P. VAN DAM, C.E. DIEZ, A. ABREU-GROBOIS & O. MCMILLAN. 2008. Dispersal, recruitment and migratory behaviour in a hawksbill sea turtle aggregation. Molecular Ecology 17: 839-853.

WALCOTT, J., S. ECKERT & J.A. HORROCKS. 2012. Tracking hawksbill sea turtles (Eretmochelys imbricata) during inter-nesting intervals around Barbados. Marine Biology 159: 927-938.