Sea Turtles

We model all three types of Argos satellite location data: LS, KF, and KS. The data are comprised of four pinnipeds, one seabird and two sea turtle species (Table 1); with deployment locations ranging between polar, temperate, and tropical marine regions (Additional file 1: Fig S1). The number of individual data sets by species and data type range from 6 to 13 with all having locations measured by GPS and at least one Argos type (Table 1). All data collected after 2008 were reprocessed by CLS to obtain the three Argos data types (4 species; Table 1).
Table 1

Number of individual data sets by species and data type. Argos data types are: Least-Squares (LS); Kalman filter (KF); Kalman smoother (KS). Mean track durations were calculated from the data after removing periods of prolonged data gaps. Tag programming details were not available for all deployments, so Argos and GPS sampling rates were calculated from the unfiltered data

Species	Common name	Code	Deployment	Mean track	Data type				GPS sample	Fastloc
			year(s)	duration (d)	LS	KF	KS	GPS	rate (min)	GPS
Zalophus californianus	California sea lion	CASL	2007	79	8	.	.	8	58	Y
Arctocephalus pusillus	Cape fur seal	CPFS	2007	25	6	.	.	6	49	Y
Dermochelys coriacea	leatherback turtle	LBTU	2008/12	92	13	.	.	13	195	Y
Eretmochelys imbricata	hawksbill turtle	HBTU	2009/10	24	6	6	6	6	124	Y
Hydrurga leptonyx	leopard seal	LESE	2018	171	8	8	8	8	47	Y
Mirounga leonina	southern elephant seal	SESE	2009/11/12/14	53	11	11	11	11	28	Y
Morus bassanus	northern gannet	NOGA	2010	23	9	9	9	9	60	N

We used an automated pre-filtering step to identify outlier observations to be ignored by the state-space model. This pre-filtering used the argosfilter R package [21 (link)] to identify locations implying travel rates >3 ms^-1 for all pinnipeds and sea turtles and travel rates >17 ms^-1 for northern gannets. These speed thresholds represent conservative upper limits of travel for these species and are intended to identify only the extreme outlier observations. This resulted in <30% of Least-Squares, <15% of Kalman filter, and <10% of Kalman smoother data being removed. The proportion of data removed by pre-filtering is considerably less than those associated with optimal speed thresholds for other species (e.g., [22 (link)]).

This final stage of the shoreline response (November 2011 and forward) defined the process whereby removal actions would be deemed complete and shoreline segments could be moved out of the response. For the first time, shoreline-oiling conditions documented by SCAT teams were compared against shoreline cleanup “endpoints,” meaning that once a segment met these final criteria, shoreline treatment was completed. As with the NFT guidelines, the SCCP endpoints were developed through consensus by representatives from the Responsible Party and Federal and State jurisdictions. The Plan included surveys of selected shoreline segments after the 2011 Atlantic hurricane season, and multiple surveys of segments post-treatment to assure that oiling conditions continued to meet endpoints. Segments that did not meet endpoints were returned to Operations for further treatment, and the inspection process was repeated.
SCAT data on oiling characteristics were used routinely to generate maps and tabular data on degree of oiling by habitat over time. Oiling degree categories (Heavy, Moderate, Light, Very Light, Trace) were defined based on the width of oiling bands on the shoreline (as measured perpendicular to the shoreline), the percent cover of oil within the band, and oil thickness using a two-step process (Figure S1 in File S1). In the first step, the width of the oil on the shoreline and the percent cover determine an initial oiling degree category; in the second step, the thickness of the oil determines the final oiling category. For example, a shoreline with a >3 m band of oil with 100% coverage is initially classified as Heavy surface cover; however, if the oil thickness is only a stain or film, the final surface oil category is Light; if the oil thickness is >0.1 cm, the final category is Heavy. The length of the shoreline is not considered in determining the degree or category of surface oiling. For example, along a marsh shoreline with highly variable orientation, there could be hundreds of meters of shoreline with no oiling then a section with tens of meters of Heavy oiling where oil stranded, adjacent to another section with Light oiling. The combination of surface oil categories and lengths of oiled shoreline provide a general level of understanding of the extent and magnitude of a spill; however, these descriptors are not adequate by themselves to develop cleanup strategies and goals for each habitat type or shoreline segment. The selection of appropriate cleanup strategies is dependent upon site-specific information regarding oiling thickness, width, distribution, and character, as well as numerous other factors including habitat condition and sensitivity, public use, wildlife use (e.g. nesting bird colonies, sea turtle nesting), and access and safety concerns.

To evaluate the conserved regions, canine BRCA2 (accession No.: NP_001006654.2), human BRCA2 (accession No.: NP_000050.3), feline BRCA2 (accession No.:NP_001009858.1), mouse BRCA2 (accession No.:NP_001074470.1), chicken BRCA2 (accession No.: NP_989607.3), sea turtle BRCA2 (accession No.: XP_037747982.1), frog BRCA2 (accession No.: ABP48763.1), gecko BRCA2 (accession No.: XP_015279682.1), and zebrafish BRCA2 (Accession No.: NP_001103864.2) amino acid sequences were obtained from the NCBI database (https://www.ncbi.nlm.nih.gov (accessed on 19 October 2021)). Multi-species alignments were performed using the multiple sequence alignment program Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/ (accessed on 19 October 2021)) according to the instruction manual (www.clustal.org/download/clustalw_help.txt, www.clustal.org/download/clustalx_help.html (accessed on 19 October 2021)) and using default settings [25 (link)], and then visualized using ESPript (http://espript.ibcp.fr/ (accessed on 28 December 2022)) [26 (link)].

Sampling methods were approved by St. George’s University Institutional Animal Care and Use Committee (IACUC-16017-R); permits were acquired from the Fisheries Division of the Grenada Government Ministry of Agriculture, Lands, Forestry, Fisheries, and the Environment.
Data for this study were collected from Levera Beach (12°13′75″ N, 61°36′78″ W) in Grenada, from the 2015 to 2019 nesting seasons from March through August (Figure 1). The beach is within a Ramsar Protected Area and is closed to public use during the leatherback nesting season. Data collection methods are those used with nest monitoring strategies employed by Ocean Spirits Inc. [4 ]. Nightly beach patrols were conducted every 30 min between 19:30 and 06:00 to detect nesting female leatherback sea turtles. Biometric data, including curved carapace length (CCL) and curved carapace width (CCW), and details from Monel (National Band and Tag Company, Kentucky, USA) and Passive Integrated Transponder (PIT) tags (Biomark, Idaho, USA) were collected to associate individual turtles with nesting data.
The 750 metre white, fine sand beach was divided into four 187 metre wide zones (zone 1–4) and subdivided with wooden markers placed 30 m apart (Zones A-Z) to allow nests to be found for excavation (Figure 1). Nest distances from the seaward edge of vegetation and high-water marks were recorded. Doomed nests laid within known washout/erosion areas, below the high-water mark, or sited too close to vegetation, were relocated to an area on the beach where the nest would be at a lower risk [4 ]. Nest relocations were dug by hand and measured using a soft tape to ensure they were of similar depth and width to a natural nest, to ensure incubation was not impacted by the relocation. Nest relocation likely influenced environmental data and overall hatchling success recorded in this project but priority was given to efforts to support conservation efforts.
From May to September of each nesting season, 10% of the total confirmed in situ and relocated nests laid at Levera Beach were selected at random and excavated, either two days post-hatchling emergence or 70 days post-ovipositioning if emergence was not observed. This delay in excavation was to provide adequate time for complete emergence of any viable hatchlings. Nest locations were excavated to a depth of one metre deep and a width of one metre wide by hand (latex gloves were worn), the excavator determining lower sand desnity by touch. Depth to the top and bottom of the egg chamber, was determined using a soft 1 m tape.
Excavated contents were catergorised based on WIDECAST (Wider Caribbean Sea Turtle Conservation Network) protcols as being: (1) empty shells, signifying hatched eggs; (2) unhatched eggs containing embryos measuring <5 mm in length to full term; (3) yolked eggs with no gross signs of development; and (4) unfertilized shelled albumin masses (Figure 2). Embryonic development success rate was calculated as the sum of the number of hatched eggs and eggs with embryos divided by the number of laid eggs (excluding shelled albumin masses), with the embryonic development success rate representing laid eggs that had the potential to develop into hatchlings. Hatchling success rate was calculated as the number of hatched eggs divided by the number of laid eggs (excluding shelled albumin masses) [25 ]. This represented the percentage of eggs that hatched. Unhatched eggs were examined for the presence of visible external or internal pink to purple discolouration, interpreted to be bacterial or fungal growth, and/or inspissation of yolk which imparted a coagulated appearance to the internal contents (Figure 2).