A species may partition its realized ecological niche along bionomic and/or scenopoetic axes due to intraspecific competition for limited resources. How partitioning manifests depends on resource needs and availability by and for the partitioning groups. Here we demonstrate the utility of analyzing short- and long-term stable carbon and nitrogen isotope ratios from imperiled marine megafauna to elucidate realized niche partitioning in these species. We captured 113 loggerhead (Caretta caretta) sea turtles at a high-use area near Crystal River, Florida, between 2016–2022, comprising 53 subadults, 10 adult males, and 50 adult females. Isotopic analyses indicated that loggerheads partition their realized ecological niche by life stage, along a scenopoetic axis. Most adults likely forage and reside in shallow areas of the study site, with most subadults displaying habitat switching between deeper locations to forage and shallower locations to rest. Some subadults, like adults, also forage and reside in shallow areas, and may display behaviors to avoid intraspecific competition. Analysis of stable isotopes with different turnover rates was critical for this characterization of intraspecific niche partitioning in loggerhead turtles, which has direct implications for ongoing research and conservation efforts for this and other imperiled species.

Capture surveys for sea turtles were conducted in the coastal waters of the Florida Big Bend between the mouths of the Crystal and Homosassa Rivers in Citrus County, Florida. Surveys were conducted approximately bi-monthly between January 2016 and February 2022 dependent on logistics and boating conditions (see Table S1). Sea turtles were spotted opportunistically from a vessel and captured using the “rodeo” method. Sighting and capture GPS locations were recorded using a Garmin GPSMAP. Captured turtles were brought on board the vessel and checked for Inconel flipper tags and passive integrated transponders (PIT). Tag numbers were recorded when already present and new flipper (National Band and Tag Company, Style 681) and PIT tags (Biomark, GPT12) were applied in the absence of old tags. Standard (nuchal notch to caudal tip) and minimum (nuchal notch to caudal notch) lengths were measured from each carapace using a tape measure (curved carapace length, CCL) and calipers (straight carapace length, SCL). Turtles sampled in this study were classified as either subadults (CCLmin < 80 cm) or adults (CCLmin > 80 cm). Turtles that were classified as adults with tails that extended past their carapace were further classified as males. All other turtles classified as adults were further classified as females. Subadults were not further classified by sex.

Epidermis and scute were sampled for isotopic analysis using 5mm and 6mm biopsy punches, respectively. Epidermis was collected from the shoulder of captured turtles after sterilization with a 95% ethanol pad and stored in dry salt (NaCl) until downstream preparation and isotopic analysis. One costal scute on each carapace was selected for keratin sampling. The selected scute was scrubbed with a Scour Pad (Scotch-Brite) and rinsed to remove small epibionts and algae. One central and one distal circular punch were made in each scute to ensure adequate sampling. Samples of all scute layers were collected, as determined by visual observation of the white carapace epithelium after removing each circular punch. Scute samples were wrapped in foil with the outer surface marked, placed in individual vials, and frozen until downstream preparation and isotopic analysis. Turtles were released within 200 m of their capture locations.

Epidermis and scute samples were sent to the Fish and Wildlife Research Institute of the Florida Fish and Wildlife Conservation Commission and the Marine Environmental Chemistry Laboratory at the University of South Florida College of Marine Science for preparation and isotopic analysis. Epidermis samples were brushed and rinsed with deionized water to remove salt, separated from any underlying fat, dried in an oven for 2 hours at 60°C, and homogenized. Lipids were extracted from epidermis and scute samples using an accelerated solvent extractor (Model 200, Dionex) with petroleum ether (3 cycles of 5 minutes heating followed by 5 minutes of static purging). 0.5–0.7 mg of each epidermis sample were weighed using a Mettler Toledo microbalance and placed into 3mm x 5mm Costech tin cups. Central and distal scute samples from each turtle were measured, and the thickest scute (representing the longest time period) from each turtle was selected for analysis. The outside layer of each scute sample was glued to a glass slide and sampled in 50 µm intervals using a carbide end mill. Each of these powdered scute layers was collected using forceps with a small piece of quartz wool, to which the scute powder is electrostatically attracted. The quartz wool and associated scute powder were then wrapped in Costech aluminum foil. Skin and scute layer samples were converted to N2 and CO2 using a Carlo – Erba EA1108 Elemental Analyzer (Thermoquest Italia). Isotope ratios and percentages-by-mass were measured in a continuous flow mass spectrometer (Delta PlusXP, Thermofinnigan, Bremen). Sample ratios are expressed in per mil (‰) as calculated using the equation:

δX = [(R_SAMPLE/R_STANDARD)-1]*1000

where X is ¹⁵N or ¹³C, and R is the ratio of ¹⁵N:¹⁴N or ¹³C:¹²C. Standards for ¹⁵N and ¹³C were AT-Air and VPDB, respectively. Secondary reference materials (NIST 8574 δ¹³C = +37.63 ± 0.10 ‰, δ¹⁵N = +47.57 ± 0.22 ‰, %N = 9.52 %, %C = 40.81 %, C:Nmolar = 5.0; NIST 8573 δ¹³C = -26.39 ± 0.09 ‰, δ¹⁵N = -4.52 ± 0.12 ‰, %N = 9.52 %, %C = 40.81 %, C:Nmolar = 5.0) were used to normalize raw measurements to standards. Measurements of analytical uncertainty (reflecting ±1 SD) were obtained by replicate measurements (n=308) of an internal laboratory reference material (NIST1577b Bovine liver, δ¹³C = -21.69 ± 0.14 ‰, δ¹⁵N = 7.83 ± 0.16 ‰, %N = 9.95 ± 0.48 %, %C =48.04 ± 0.71 %, C:Nmolar = 5.63 ± 0.27), and were ≤ 0.25 ‰ (δ¹³C) and ≤ 0.31 ‰ (δ¹⁵N).